Hepatobiliary scintigraphy (HBS) is frequently obtained for evaluation of patients with symptoms concerning for recurrent biliary disease. Although a depressed gallbladder (GB) ejection fraction (GBEF) is generally considered the most accurate and common scintigraphic means for identifying chronic GB disease (CGBD), other secondary scintigraphic findings have long been considered somewhat specific for the presence of chronic cholecystitis including delayed small bowel (SB) transit, slow GB filling, and reversal of the normal GB followed by SB filling sequence.1–8 Many of these studies demonstrated limitations, including small study sizes, subjects with a very high prevalence of gallstones, use of ultrasound and/or oral cholecystography as the gold standard for disease diagnosis, or application of varying cholecystokinin (CCK) administration protocols that have since been shown to have poor specificity. These factors bring into question the generalizability of such associations with CGBD.
In today’s typical medical practice, patients suspected of having CGBD undergo ultrasound (US) evaluation.9 If US is positive for features of chronic cholecystitis (eg, gallstones and/or wall thickening), patients are frequently offered laparoscopic cholecystectomy without further imaging. In cases where US findings are negative or equivocal, patients often undergo further evaluation with HBS for a determination of GBEF.10–12 After appropriate administration of CCK, a GBEF of 38% or greater is considered normal, and a GBEF less than 38% equates with a diagnosis of CGBD.7 Given this modern paradigm for the evaluation and management of patients with CGBD, we hypothesize that the aforementioned classical HBS findings for chronic cholecystitis are not applicable to today’s scintigraphic diagnosis of CGBD.
Chronic GB disease may be designated by a variety of terms including biliary dyskinesia, chronic cholecystitis, chronic acalculous cholecystitis, acalculous biliary disease, acalculous cholecystopathy, and functional GB disorder.7,9 These terms are often used interchangeably, leading to confusion. The association between a low GBEF and many forms of CGBD has been well documented; however, the secondary, classical HBS findings discussed previously have been studied predominantly in the setting of chronic calculous cholecystitis.1,2,7,8,13
The primary objective of this study was to determine if the scintigraphic findings of delayed SB transit (>60 minutes), slow filling of the GB (late in the first hour of dynamic imaging), reversal of the normal GB and SB filling sequence (ie, GB visualized after SB), or the difference in time between GB fill and SB transit are associated with CGBD. The secondary objectives were to determine if the GBEF correlated with GB fill time, SB transit time, or the difference between GB fill and SB transit times. If these classical findings were to correlate with CGBD, it would be possible to forgo the additional hour of imaging and CCK administration necessary for GBEF determination.
PATIENTS AND METHODS
Institutional review board approval was obtained for the study. Written informed consent was waived, given the retrospective nature of the study.
We reviewed all HBS exams performed at our institution from January 2014 through April 2015, totaling 366 studies (Fig. 1). We included HBS exams that used the 1-hour CCK infusion protocol for GBEF determination and evaluation of symptoms concerning for chronic biliary disease. We excluded patients with factors known to alter normal biliary flow, including a history of previous biliary procedure (eg, sphincterotomy or common bile duct stenting), recent use of morphine, and/or pretreatment with CCK.5,14 Of the 366 exams, 126 did not meet inclusion criteria because of acute biliary disease, such as acute cholecystitis or biliary leak. An additional patient was not included because of spontaneous GB contraction without CCK infusion. Of the 239 exams meeting inclusion criteria, one was excluded for having a biliary stent, one was excluded for having received a dose of morphine shortly before HBS, and 16 others were excluded for having been pretreated with CCK. The remaining 221 HBS studies were included in our final analysis (Fig. 1).
The mean patient age was 45.3 ± 15.2 years (range, 15–87); 152 were female subjects (69%), and 69 were male subjects (31%) (Table 1). Four (2%) were pediatric patients, ages 15 to 17 years; 16 (7%) of the exams were obtained on patients in the emergency department or inpatient setting. The remaining 205 (93%) exams were performed on an outpatient basis. In our analysis, we included US studies performed before or within 6 months after the HBS study. Approximately 187 (85%) of the patients had an US of the right upper quadrant (Table 1). Of these, 10 (5%) demonstrated wall thickening greater than 3 mm, and 21 (11%) demonstrated gallstones within the GB (Table 1). We also performed a search of our institution’s medical record system to determine how many patients underwent surgery after the hepatobiliary scan. Approximately 49 (22%) patients went on to have a cholecystectomy (Table 1). Of the 49 patients, 13 had abnormal GBEF values, and 36 had normal GBEF values on HBS exam (data not shown). The surgical pathology demonstrated chronic cholecystitis in 12 of the 13 patients with abnormal GBEF values and 33 of the 36 patients with normal GBEF values (data not shown).
All 205 outpatient exams fasted overnight before the morning HBS exam, typically performed at 8 am. For the 16 exams performed on inpatients or those from the ED, patients fasted for a minimum of 6 hours but no more than 24 hours before the beginning of the study. Adult patients were administered a dose of 5 mCi (mean, 5.3 mCi; range, 4.5–5.5 mCi) of 99mTc mebrofenin IV, and pediatric patients were administered a weight-based dose in accordance with the European Association of Nuclear Medicine guidelines (http://www.eanm.org/publications/dosage-calculator/) for radiopharmaceutical dosages (mean, 3.5 mCi; range, 2.9–3.7 mCi). Immediately after injection, dynamic imaging was obtained in the anterior projection for 1 hour at 1 frame per minute to evaluate blood pool clearance of radiotracer, hepatic uptake, excretion into the biliary tree, GB filling, and biliary-to-bowel transit. At the end of 1 hour, static left anterior oblique and right lateral views were obtained for confirmation of GB filling. The patients were then promptly given 0.02 μg/kg CCK (mean, 1.6 mg; range, 1.0–3.3 mg for adults; mean, 0.9; range, 0.8–0.9 mg for pediatrics) in 100 mL of normal saline, followed by additional dynamic imaging in the left anterior oblique projection for 1 hour during CCK administration for the determination of GBEF in accordance with consensus recommendations outlined by DiBaise et al,10 Society of Nuclear Medicine and Molecular Imaging practice guidelines,11 and American College of Radiology–Society for Pediatric Radiology practice parameters.12
Gallbladder filling time, SB transit time, sequence of GB and SB filling, and GBEF were recorded for each patient (Figs. 2–6). Cases with delayed SB transit proceeded to GBEF determination with prompt CCK administration. In these 44 cases, SB transit was seen immediately after CCK administration and estimated at 65 minutes (based on average time delay of 5 minutes for administration of CCK). This artificially underestimated the SB transit time. Gallbladder ejection fraction was calculated by placing a region of interest over the GB immediately before administration of CCK and at the end of CCK infusion, taking into account background activity, patient motion, and potential superimposed radiotracer activity in the bowel. All regions of interest and time activity curves for GB emptying were reviewed by a nuclear medicine physician for appropriate positioning and absence of artifacts or other factors which could degrade accuracy. Gallbladder ejection fraction was calculated using the below formula, as detailed in the aforementioned practice guidelines/parameters and consensus recommendations.
To quantify the difference between SB transit time and GB fill time, we defined a new measurement, which we termed transit differential (transit differential = SB transit time − GB fill time). Positive transit differential values indicate a normal filling sequence (ie, GB filling followed by SB transit), and negative transit differential values indicate a reversal of normal filling sequence (ie, SB transit followed by GB filling). Comparative analysis of age, sex, GB fill time, normal SB transit, reversal of normal GB and SB filling sequence, and transit differential for normal and abnormal GBEF values was performed with the t test for continuous variables and the Fisher test for nominal variables. P < 0.05 was considered statistically significant. Correlation analysis was performed using linear regression techniques comparing GBEF values and GB fill times, SB transit times, and transit differentials. Correlation coefficients (R2) were determined, with R2 = 0 indicating no correlation and R2 = 1 indicating complete correlation. R2 values of at least 0.3 to 0.4 are required for mild clinical correlation. JMP software (SAS, Raleigh, NC) was used for the statistical analyses.
Of the 221 patients studied, 201 (91%) had normal GBEF values (≥38%), and 20 (9%) had an abnormally low GBEF values (<38%) (Table 2). The mean GBEF was 78% ± 24% (average ± standard deviation) (Table 2). The mean GBEF for patients with a normal ejection fraction value (≥38%) was 84% ± 16%, and that for patients with an abnormal GBEF value (<38%) was 21% ± 11% (Table 3). The mean GB fill time for all 221 patients was 13 ± 10 minutes (Table 2). For patients with a normal GBEF value, the GB fill time was 13 ± 10 minutes, and for patients with an abnormal GBEF value, the mean GB fill time was 15 ± 10 minutes. All of the patients with normal and abnormal GBEF values had normal GB fill times.
Of the 221 patients, 177 (80%) had normal SB transit (≤60 minutes) with a mean SB transit time of 31 ± 21 minutes (Table 2). The remaining 44 patients had SB transit times exceeding 60 minutes. In each instance, the SB transit occurred immediately after commencement of CCK administration. Approximately 159 (79%) of the 201 patients with a normal GBEF value had a normal SB transit time with mean SB transit time of 31 ± 21 minutes (Table 3). Approximately 18 (90%) of the 20 patients with an abnormal GBEF value had normal SB transit with a mean SB transit time of 27 ± 18 minutes (Table 3).
Of the 221 patients, 167 (76%) had normal GB/SB filling sequence (GB filling followed by SB transit) (Table 2). Approximately 153 (77%) patients with normal GBEF values had normal GB/SB filling sequence, and 48 (23%) had reversal of normal GB/SB filling sequence (Table 3). Approximately 14 (70%) patients with abnormal GBEF values had normal GBSB filling sequence, and 6 (30%) had reversal of normal GB/SB filling sequence (Table 3). Mean transit differential for all 221 patients was 17 ± 26 minutes (Table 2). Mean transit differential for patients with normal GBEF values was 18 minutes, and that for those with abnormal GBEF values was 12 minutes (Table 3).
Our results demonstrated no significant difference (P > 0.05) between groups with normal and abnormal GBEF values based on age, sex, GB fill time, normal SB transit, reversal of normal GB and SB filling sequence, and transit differential (Tables 2 and 3). Additionally, our results demonstrated no significant correlation between the GBEF and GB filling time (Fig. 7), SB transit time (Fig. 8), and transit differential (Fig. 9).
The scintigraphic findings of delayed SB transit, delayed GB filling, and reversal of the normal GB/SB filling sequence have traditionally been associated with chronic biliary disease. The purpose of this study was to assess whether these classically associated findings can be used for the diagnosis of CGBD, when compared with GBEF. We did not find any association between GBEF and these classically reported findings. Therefore, we propose that these classical HBS findings should not be used for CGBD diagnosis, and GBEF determination should continue to be used as the gold standard.
The association between delayed SB transit time and chronic cholecystitis was initially shown by Weissmann in her seminal work presented at RSNA in 1978.13 In her study of 51 patients with delayed SB transit (>60 minutes) on HBS performed with 99mTc IDA, 33 individuals (65% of cases) had a concurrent diagnosis of chronic cholecystitis, 2 had acute cholecystitis, and the remaining 16 had normal GBs. The likelihood of the association between delayed SB transit time and chronic cholecystitis was shown to increase as transit times increased. There was a 50% association (8 of 16 cases) when transit occurred near 1 hour, 67% (10/15) association when transit occurred in 1 to 2 hours, and an 85% association (17/20) when SB transit was 2 hours or longer (3). This same work by Weissmann also demonstrated an association between delayed GB filling and chronic cholecystitis.13 Approximately 37 patients (73%) with delayed GB filling on HBS (>60 minutes) had chronic cholecystitis. As demonstrated with SB transit, increasing delays in GB filling correlated with an increasing likelihood of chronic cholecystitis (100% of patients with GB filling greater than 1.5 hours had chronic cholecystitis).
Al-Sheikh et al were the first to demonstrate that reversal of normal GB/SB filling was associated with chronic cholecystitis.2 In their study, they evaluated 134 patients with symptoms of CGBD with HBS using 5 mCi of 99mTc PIPIDA. A “normal” HBS exam result was defined as filling of both the GB and SB by 1 hour regardless of filling sequence (ie, no acute cholecystitis or evidence of biliary obstruction). They also evaluated all patients with US and/or oral cholecystography, diagnosing chronic cholecystitis based on the combination of symptoms and presence of gallstones. Using these criteria, 134 normal HBS exams were analyzed for the sequence of GB filling and SB transit. Approximately 102 cases had filling of the GB before SB transit, whereas 32 demonstrated reversal of this sequence. Of the 32 patients with reversal of filling, 24 (75%) were diagnosed with chronic cholecystitis. This contrasted with 29 (28%) of 102 patients with normal fill sequence having chronic cholecystitis. Based on their results, reversal of normal GB/SB filling had a specificity of 90%, an accuracy of 73%, and a positive predictive value of 75% for chronic cholecystitis, whereas a normal fill sequence had a negative predictive value of 72%.
The association between GB/SB filling sequence reversal and chronic cholecystitis was again demonstrated by Achong and Oates using the parameters defined by Al-Sheikh et al, with the exception of a different radiotracer (99mTc mebrofenin).1,2 They analyzed 141 “normal” HBS exam results, correlating all exams with ultrasound findings, and a subset of patients (35) with histopathology. Using US as the gold standard, the diagnosis of chronic cholecystitis was based on the sonographic findings of gallstones/sludge, wall thickening, and/or nonvisualization of the GB. Approximately 81 of the 141 cases had chronic cholecystitis (69 chronic calculous cholecystitis and 12 chronic acalculous cholecystitis); 48 had reversal of normal GB and SB filling sequence. Reversal of the filling sequence had a specificity of 79% for chronic cholecystitis when diagnosed with US and a specificity of 100% when confirmed histopathologically.
In each of these studies, there was a high prevalence of gallstones underlying the diagnosis of chronic cholecystitis. The work by Achong and Oates was the only study to specifically note the presence of chronic acalculous cholecystitis.1 The presence of gallstones is hypothesized to contribute to the delay in GB filling, the delay in SB transit, and the reversal of normal GB/SB filling sequence. Recurrent passage of calculi leads to inflammation of the cystic duct and common bile duct reversing the normal pressure differential necessary for GB fill and biliary-to-bowel transit.15 The obstruction of physiologic flow leads to stasis of abnormally viscous bile, further impeding the retrograde inflow of radiolabeled bile.
In contradistinction to these studies, our results demonstrate no significant association between the scintigraphic findings of delayed SB transit, slow GB filling, or reversal of the normal GB/SB filling sequence and CGBD as diagnosed by GBEF. The discrepancy between our findings and those of previous studies may in part be explained by our larger sample size, our use of GBEF as the gold standard for CGBD diagnosis, current HBS practice guidelines,11,12 a lower prevalence of gallstones in the current patient population undergoing HBS exam, and current medical treatment and diagnosis algorithms. Within the current HBS practice guidelines, mebrofenin is the preferred radiotracer, whereas previous studies used IDA and PIPIDA. The differences in extraction efficiency and other factors between these radiopharmaceuticals may account for some of the variation in results. Additionally, at our institution, most patients with suspected biliary disease undergo initial evaluation by US. If US exams demonstrate gallstones or other evidence of chronic cholecystitis, such patients are most commonly offered cholecystectomy, without previous HBS. In most circumstances, our patients only undergo HBS if US results are equivocal or negative, and there is persistent clinical suspicion that they have CGBD. Given these differences and our data which show no association between GBEF and classical HBS findings, we propose that these classical findings should not be used for the diagnosis of CBGD in modern nuclear medicine practice.
We recognize limitations to our research, including a small number of patients with abnormal GBEF values (9% of cases). This may be explained by the possibility that many patients who might have been referred for HBS underwent surgery. The retrospective nature of the study raises the possibility of selection bias and/or confounding. The absolute gold standard for diagnosis is histopathology; however, most of these patients do not undergo surgery, making GBEF a reasonable diagnostic alternative. It is important to note that GBEF can be used for the diagnosis of CGBD but cannot distinguish among the various etiologies of chronic biliary disease. A final limitation is our underestimation of true SB transit time when SB activity was not seen within the first hour of dynamic imaging. In these situations, CCK stimulation commenced promptly, and radiotracer was immediately seen in the duodenum in all cases. Therefore, an SB transit value of 65 minutes was estimated. This likely resulted in a decreased SB transit and transit differential value for 44 HBS exams.
Delayed SB transit, slow GB filling, and reversal of the normal GB and SB filling sequence have classically been associated with chronic cholecystitis. Our research demonstrates that these classical scintigraphic findings do not correlate with CGBD as diagnosed by the presence of abnormal GBEF values. Thus, these findings are of doubtful diagnostic significance and should not be used as surrogate markers for CGBD diagnosis in today’s medical practice. Gallbladder ejection fraction should continue to be used as a diagnostic tool for CGBD in HBS.
1. Achong D, Oates E. A reversed sequence of gallbladder and small bowel visualization during cholescintigraphy. Its relationship to chronic cholecystitis
. Clin Nucl Med
2. Al-Sheikh W, Hourani M, Barkin JS, et al. A sign of symptomatic chronic cholecystitis
on biliary scintigraphy. Am J Roentgenol
3. Chamarthy M, Freeman LM. Hepatobiliary scan findings in chronic cholecystitis
. Clin Nucl Med
4. DeCamp J, Tabatowski K, Schauwecker D, et al. Comparison of gallbladder ejection fraction
with histopathologic changes in acalculous biliary disease. Clin Nucl Med
5. Kim CK, Palestro CJ, Solomon RW, et al. Delayed biliary-to-bowel transit in cholescintigraphy after cholecystokinin treatment. Radiology
6. Krishnamurthy GT, Krishnamurthy S. Extended application of 99m
Tc-mebrofenin cholescintigraphy with cholecystokinin in the evaluation of abdominal pain of hepatobiliary and gastrointestinal origin. Nucl Med Commun
7. Tulchinsky M. Applied hepatobiliary scintigraphy
in chronic gallbladder diseases. Appl Radiol
8. Ziessman HA. Functional hepatobiliary disease: chronic acalculous gallbladder and chronic acalculous biliary disease. Semin Nucl Med
9. Behar J. Physiology and pathophysiology of the biliary tract: the gallbladder and sphincter of Oddi—a review. ISRN Physiol
. 2013;2013:1–15. Available at: https://www.hindawi.com/journals/isrn/2013/837630/
10. DiBaise JK, Richmond BK, Ziessman HH, et al. Cholecystokinin-cholescintigraphy in adults: consensus recommendations of an interdisciplinary panel. Clin Gastroenterol Hepatol
. Elsevier. 2011;9:376–384.
11. Tulchinsky M, Ciak BW, Delbeke D, et al. SNM practice guideline for hepatobiliary scintigraphy
4.0. J Nucl Med Technol
12. ACR-SPR. Practice Parameter Hepatobiliary Scintigraphy [Internet]
. 2017; Available at: https://www.acr.org/-/media/ACR/Files/Practice-Parameters/Hepato-Scint.pdf
. Accessed August 8, 2018.
13. Weissmann HS. Freeman and Johnson’s Clinical Radionuclide Imaging
. 3rd ed. Freeman LM, editor. Orlando, FL: Grune and Stratton, Inc; 1986.
14. Jaffe JH, Martin WA. The Pharmacological Basis of Therapeutics
. 5th ed. Goodman LS, Louis S, Gilman A, editors. New York, NY: Macmillan; 1975.
15. Krishnamurthy GT, Bobba VR, McConnell D, et al. Technique quantitative biliary dynamics: introduction of a new noninvasive scintigraphic. J Nucl Med J Nucl Med