- Email: [email protected]
Acoustic Radiation Force Impulse Imaging of the Pancreas for Estimation of Pathologic Fibrosis and Risk of Postoperative Pancreatic Fistula
Acoustic Radiation Force Impulse Imaging of the Pancreas for Estimation of Pathologic Fibrosis and Risk of Postoperative Pancreatic Fistula Nobuhiro Harada, MD, Takeaki Ishizawa, MD, PhD, Yosuke Inoue, MD, PhD, Taku Aoki, MD, PhD, Yoshihiro Sakamoto, MD, PhD, Kiyoshi Hasegawa, MD, PhD, Yasuhiko Sugawara, MD, PhD, Mariko Tanaka, MD, PhD, Masashi Fukayama, MD, PhD, Norihiro Kokudo, MD, PhD We sought to evaluate whether pancreatic elasticity, measured using acoustic radiation force impulse (ARFI) imaging, can determine the degree of pancreatic fibrosis and risk of pancreatic fistula (PF) in patients undergoing pancreatic resection. Although soft pancreatic texture is a reliable predictor of postoperative PF, noninvasive, quantitative methods of assessing pancreatic hardness have not been established. STUDY DESIGN: Shear wave velocity (SWV) of the pancreas was preoperatively measured by ARFI imaging in 62 patients undergoing pancreatic resection. Correlations of SWV with pathologic degree of fibrosis in the resected pancreas, exocrine function of the remnant pancreas, and the incidence of postoperative PF were determined. RESULTS: The SWV was positively correlated with the degree of pancreatic fibrosis (Spearman’s rank correlation coefficient [r] ¼ 0.660, p < 0.001) and inversely correlated with postoperative amylase concentrations and daily output of pancreatic juice. The incidence of postoperative PF was significantly higher in the 32 patients with soft (SWV < 1.54 m/s) than in the 30 with hard (SWV 1.54 m/s) pancreata (63% vs 17%, p < 0.001). Multivariate analysis showed that a soft pancreas (SWV < 1.54 m/s) was an independent predictor of postoperative PF (odds ratio 38.3; 95% CI 5.82 to 445; p ¼ 0.001). CONCLUSIONS: Pancreatic elasticity on preoperative ARFI imaging accurately reflected the pathologic degree of fibrosis and exocrine function of the pancreas, enabling surgeons to adopt appropriate surgical procedures according to the risk of postoperative PF in each patient undergoing pancreatic resection. (J Am Coll Surg 2014;219:887e894. 2014 by the American College of Surgeons)
BACKGROUND:
as severe infections and massive hemorrhage, it is important to accurately estimate the risk of PF in individual patients undergoing pancreatic resection. Several clinical factors were found to be significant predictors of postoperative PF, including soft pancreatic texture,2,6-11 small main pancreatic duct,11-15 tumor pathology other than adenocarcinoma,15,16 and larger amounts of intraoperative blood loss.6,15,17 Despite its subjectivity, hardness of the pancreas by palpation during surgery is regarded as a simple and reliable indicator for the development of postoperative PF. Quantitative methods are needed to assess the hardness of pancreatic tissue and to determine whether hardness correlates with the degree of pathologic fibrosis. Several ultrasound elastographic techniques were recently developed for noninvasive evaluation of tissue elasticity.18,19 One method, acoustic radiation force impulse (ARFI) imaging, measures the velocity of shear waves generated by tissue after irradiation with
Despite recent advances in surgical techniques and perioperative management of pancreatic resection, postoperative pancreatic fistula (PF) remains a major postoperative concern, with an incidence of 10% to 50%.1-5 Because postoperative PF can lead to life-threatening events such Disclosure Information: Nothing to disclose. Support: This work was supported by grants from the Takeda Science Foundation; the Kanae Foundation for the Promotion of Medical Science; the Ministry of Education, Culture, Sports, Science and Technology of Japan; and the Ministry of Health, Labor and Welfare of Japan. Received May 29, 2014; Revised July 17, 2014; Accepted July 30, 2014. From the Hepato-Biliary-Pancreatic Surgery Division, Department of Surgery (Harada, Ishizawa, Inoue, Aoki, Sakamoto, Hasegawa, Sugawara, Kokudo), and the Department of Pathology (Tanaka, Fukayama), Graduate School of Medicine, The University of Tokyo, Tokyo, Japan. Correspondence address: Norihiro Kokudo, MD, PhD, Hepato-BiliaryPancreatic Surgery Division, Department of Surgery, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan. email: [email protected]
ª 2014 by the American College of Surgeons Published by Elsevier Inc.
887
http://dx.doi.org/10.1016/j.jamcollsurg.2014.07.940 ISSN 1072-7515/14
888
Harada et al
Preoperative Pancreatic Elasticity Assessment
Abbreviations and Acronyms
ARFI DP MPD OR PD PF POD ROC ROI SWV
¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼
acoustic radiation force impulse distal pancreatectomy main pancreatic duct odds ratio pancreatoduodenectomy pancreatic fistula postoperative day receiver operating characteristic region of interest shear wave velocity
short-duration acoustic pulses and enables evaluation of tissue elasticity in a small (<1.0 cm) region of interest (ROI) on conventional B-mode ultrasonographic images.20,21 Using this technique, shear wave velocity (SWV) correlates with tissue hardness. In clinical settings, ARFI imaging has been used mainly to estimate the degree of hepatic fibrosis without biopsy in patients with chronic liver disease,19,22-24 and has been used for the differential diagnosis of nodular lesions in the liver,25 breast,26 thyroid,27 and pancreas.28 To our knowledge, however, ARFI imaging has not been used to estimate pancreatic hardness or whether the latter predicts a risk of PF after pancreatic surgery, except in the recent preliminary study by Lee and colleagues.29 The aim of this study was to evaluate whether the SWV of the pancreas, as measured by preoperative ARFI imaging, correlates with the pathologic degree of pancreatic fibrosis and pancreatic exocrine function in patients undergoing pancreatic resection. The ability of “soft” vs “hard” pancreata, based on preoperative SWV, to predict postoperative PF was also assessed.
J Am Coll Surg
METHODS This study was conducted with the approval of the Institutional Ethics Review Board of our institution. All patients provided written informed consent. Patients The study cohort consisted of 68 patients who underwent pancreatic resection at the University of Tokyo Hospital between February 2012 and February 2013. Preoperative ARFI imaging could not be performed in 6 of these patients because of difficulties obtaining clear and static images of the pancreatic parenchyma. In the remaining 62 patients, the SWV of the pancreas was measured by ARFI imaging before pancreatoduodenectomy (PD, n ¼ 35) or distal pancreatectomy (DP, n ¼ 27), with these SWVs compared with postoperative pancreatic exocrine function and the incidence of PF. Pancreatoduodenectomy was performed as previously described.30,31 Briefly, the pancreatic parenchyma was divided using the clamp crushing method, with meticulous ligation of small vessels. Two Roux limbs were created and used for digestive reconstruction: one for biliary and pancreatic anastomoses, and the other for gastrojejunostomy. Patients with a narrow main pancreatic duct (MPD) and/or a soft pancreatic texture, as assessed by manual palpation, underwent PD without pancreatojejunostomy, ie, totally external tube pancreatostomy, followed 3 months later by 2stage pancreatojejunostomy.30 Other patients underwent pancreatojejunostomy with duct-to-mucosa anastomosis and external tube pancreatostomy during PD. Distal pancreatectomy was also performed by dividing the pancreatic parenchyma using the clamp crushing
Figure 1. Preoperative acoustic radiation force impulse imaging of the pancreas and pathologic images of the resected specimen. (A) The shear wave velocity (SWV) of the pancreas was measured by setting a region of interest (ROI) in the pancreatic body above the confluence of the splenic and superior mesenteric veins (yellow arrows), away from the pancreatic tumor (arrowheads in left) and the superior mesenteric artery (yellow arrowhead in right). (B) Pathologic pancreatic fibrosis grading according to Klo¨ppel’s criteria. The SWVs were 1.01 m/s for normal fibrosis, 1.35 m/s for mild fibrosis, 1.70 m/s for moderate fibrosis, and 1.81 m/s for severe fibrosis.
Vol. 219, No. 5, November 2014
Harada et al
Table 1. Relationships Between Pancreatic Shear Wave Velocity and Patient Demographic and Clinical Characteristics Variables
Age, y >70 70 Sex Male Female MPD diameter, mm >3 3 Body mass index, kg/m2 >25.0 25.0 Diabetes mellitus Present Absent Pathologic diagnosis Pancreatic cancer Other diseases*
n
SWV, m/s
p Value
30 32
1.77 (1.01e3.52) 1.45 (0.85e2.37)
0.041
38 24
1.68 (0.94e3.52) 1.49 (0.85e2.50)
0.435
34 28
1.78 (1.14e3.52) 1.40 (0.85e3.19)
0.001
50 12
1.61 (0.85e3.52) 1.59 (0.94e2.37)
0.908
28 34
1.68 (0.95e3.52) 1.55 (0.85e2.98)
0.392
31 31
1.71 (0.94e3.52) 1.50 (0.85e2.98)
0.078
Continuous data are shown as median (range). *Including intraductal papillary mucinous neoplasm (n ¼ 9); carcinoma of the papilla of Vater (n ¼ 5), the lower bile duct (n ¼ 4), and the duodenum (n ¼ 2); pancreatic neuroendocrine tumor (n ¼ 5), gastric cancer (n ¼ 2), serous cystic neoplasm (n ¼ 2), solid-pseudopapillary neoplasm (n ¼ 1), and chronic pancreatitis (n ¼ 1). MPD, main pancreatic duct; SWV, shear wave velocity.
method and by ligating the small vessels and the MPD.32 Prophylactic abdominal drains were routinely placed on the superior and inferior aspects of the pancreatic stump. Preoperative acoustic radiation force impulse imaging of the pancreas Preoperative ARFI imaging was performed using the ACUSON S2000 (Siemens Medical Solutions) in the Virtual Touch Tissue Quantification mode, as described.19-29,33 Briefly, patients in the fasting state were placed in the decubitus position. To accurately measure SWVs, conventional B-mode ultrasonography was performed to visualize the pancreatic parenchyma within a depth of 8 cm from the transducer.21,33 An ROI measuring 10 mm axially by 6 mm in width was set in the pancreatic body above the confluence of the splenic and portal veins (splenoportal junction), away from the pancreatic tumors and the superior mesenteric artery (Fig. 1A). Before using preoperative ARFI imaging to evaluate the study population, the reproducibility of this technique was validated in 3 healthy volunteers (3 men, ages 24, 31, and 38 years) and in 3 patients (2 men
Preoperative Pancreatic Elasticity Assessment
889
and 1 woman, ages 71, 72, and 82 years) who underwent PD for biliary or pancreatic malignancies. Three surgeons, blinded to results obtained by the others, measured SWVs 10 times in each subject. In this validation trial, pancreatic SWVs measured by the 3 surgeons did not differ significantly, either in the group of healthy volunteers or in the group of patients undergoing PD (ANOVA, Appendix 1: Supplementary Tables 1 to 5, online only). Shear wave velocities in the study population were measured 5 times in each of the 62 patients; the mean SWV of the median 3 values for each patient was calculated and used in further analyses. Pathologic assessment of the pancreas Pancreatic tissue samples were obtained 0.5 to 1.0 cm from the pancreatic stump, fixed in formalin, and stained with hematoxylin and eosin. These slides were evaluated by an experienced pathologist (MT), blinded to each patient’s clinical data and postoperative course. The degree of pancreatic fibrosis was scored according to perilobular and intralobular fibrosis and divided into 4 categories of Klo¨ppel scores: 1 to 3, normal tissue; 4 to 6, mild fibrosis; 7 to 9, moderate fibrosis; and 10 to 12, severe fibrosis (Fig. 1B).34 Fatty infiltration into pancreatic tissue was evaluated microscopically as 1 to 2, no fatty infiltration or 3 to 4, fatty infiltration.10 Based on the results of SWVs and pancreatic fibrosis score, cut-off values of SWV predicting pathologic pancreatic fibrosis, defined as a Klo¨ppel score 7, were calculated using a receiver operating characteristic (ROC) curve and a common optimization step maximizing Youden’s index.35 Evaluation of pancreatic exocrine function and postoperative pancreatic fistula As indicators of pancreatic exocrine function, amylase concentrations and daily amounts of pancreatic juice were evaluated using fluid samples drained from an external pancreatic tube in 35 patients who underwent PD. The amylase concentrations in pancreatic juice were not evaluated in 5 patients; 2 showed no discharge of pancreatic juice after surgery, and amylase concentrations were not determined in the other 3. Postoperative PF was defined as an amylase concentration in fluid more than 3 times the upper limit of serum amylase concentration after the third postoperative day.5 Statistical analyses Continuous data were expressed as median (range). Quantitative and categorized variables were compared using the Wilcoxon rank-sum test and Fisher’s exact test, respectively. The correlations of SWV with fibrosis score, fatty
890
Harada et al
Preoperative Pancreatic Elasticity Assessment
J Am Coll Surg
Figure 2. Correlations between pancreatic shear wave velocity and pathologic findings of resected specimens. The preoperative shear wave velocity (SWV) of the pancreas was positively correlated with (A) pathologic fibrosis score (Spearman’s rank correlation coefficient [r] ¼ 0.660, p < 0.001), but not with (B) pathologic fatty infiltration score (r ¼ 0.058, p ¼ 0.654).
infiltration score, and pancreatic exocrine function were evaluated using Spearman’s rank correlation test. Multivariate analysis was performed to evaluate the ability of preoperative SWV to predict the risk of postoperative PF. Due to their clinical significance, 8 potentially important preoperative and intraoperative factors were included in the analysis: age (70 years vs >70 years),36 sex,6 diabetes mellitus (yes vs no), body mass index (25 kg/m2 vs >25 kg/m2),6,37 pathology of the tumor (pancreatic cancer vs other neoplasm),15,16 amount of blood loss (1,000 mL vs >1,000 mL),6,15,17 MPD diameter (3 mm vs >3 mm),12,13 and SWV (1.54 m/s vs >1.54 m/s, based on ROC curve analysis). The results were expressed as adjusted odds ratios (OR) with 95% confidence intervals (CI), and p values were calculated using the likelihood ratio test. Values of p < 0.05 were considered statistically significant. All statistical analyses were performed using JMP software version 9.0.0 (SAS Institute).
RESULTS Shear wave velocities of the pancreas and patients’ background characteristics The study subjects consisted of 38 men and 24 women, ranging in age from 28 to 85 years (median 70 years).
Median preoperative pancreatic SWV was 1.49 m/s (range 0.85 to 3.52 m/s). Preoperative SWVs of the pancreas were higher in patients aged 70 years or younger and in patients with an MPD diameter of 3 mm or less (Table 1). The median time to measure SWV of the pancreas was 157 seconds (range 42 to 490 seconds). Shear wave velocities of the pancreas and pathologic diagnosis of fibrosis and fatty infiltration Preoperative SWV of the pancreas significantly correlated with pathologic fibrosis score (Spearman’s rank correlation coefficient [r] ¼ 0.660, p < 0.001; Fig. 2A), but not with pathologic fatty infiltration score (r ¼ 0.058, p ¼ 0.654; Fig. 2B). Receiver operating characteristic curve analysis was performed to determine a cut-off value of SWV predictive of pancreatic fibrosis, defined as a Klo¨ppel score 7 (Fig. 3). An SWV cut-off of 1.54 m/s yielded an area under the curve of 0.881 (p < 0.001). The sensitivity, specificity, positive predictive value, and negative predictive value of this cut-off were 91%, 75%, 67%, and 93%, respectively. Based on these results, the 62 patients enrolled in this study were divided into soft (SWVs < 1.54 m/s, n ¼ 32) and hard (SWV 1.54 m/s, n ¼ 30) pancreas groups in the subsequent analyses.
Vol. 219, No. 5, November 2014
Harada et al
Preoperative Pancreatic Elasticity Assessment
891
and 71% (5 of 7) in the patients who had undergone primary and secondary pancreatojejunostomy. The incidence of postoperative PF (63% vs 17%, p ¼ 0.001) and the morbidity rate, defined as Clavien-Dindo grade III or higher38 (28% vs 10%, p ¼ 0.108) were significantly higher in the soft than in the hard pancreas group (Table 2). Multivariate analysis revealed that soft pancreas, defined as a preoperative SWV < 1.54 m/s, was an independent predictive factor of postoperative PF (odd ratio [OR] 38.3, 95% CI 5.82 to 445; p ¼ 0.001). Other independent predictors of postoperative PF were age 70 years (OR 7.69, 95% CI 1.44 to 62.3; p ¼ 0.030) and male sex (OR 8.87, 95% CI 1.71 to 67.8; p ¼ 0.018, Table 3).
Figure 3. Receiver operating characteristic curves analysis of shear wave velocity for predicting pathologic fibrosis of the pancreas. Receiver operating characteristic curve analysis was used to determine the cut-off value of shear wave velocity (SWV) for predicting pancreatic fibrosis, defined as a Klo¨ppel score 7. When the SWV cut-off was set at 1.54 m/s, it provided an area under the curve of 0.881 (p < 0.001) with sensitivity, specificity, positive predictive value, and negative predictive value of 91%, 75%, 67%, and 93%, respectively.
Shear wave velocities of the pancreas and pancreatic exocrine function In the 35 patients who underwent PD, higher preoperative SWV of the pancreas was significantly correlated with a lower volume of pancreatic juice on postoperative day (POD) 7 (Fig. 4A) and lower maximum amylase concentrations in pancreatic juice on PODs 1 and 3 (Fig. 4B). Of the 35 patients who underwent PD, 23 had a hard and 12 had a soft pancreas, with the former having significantly lower median volumes of pancreatic juice on POD 7 (40 mL [range 0 to 155 mL] vs (215 mL [range 130 to 420 mL], p < 0.001) and significantly lower median maximum amylase concentrations in pancreatic juice on PODs 1 to 3 (3,062 units/L [range 2 to 16,350 units/L]) vs 22,747 units/L [range 8,154 to 54,270 units/L], p < 0.001). Shear wave velocity-based hardness of the pancreas and operative outcomes Postoperative PF (International Study Group on Pancreatic Fistula [ISGPF] grade B or higher) developed in 12 of the 35 patients who underwent PD (34%) and 27 of those who underwent DP (44%). Among the 35 PD patients, the incidences of postoperative PF were 25% (7 of 28)
DISCUSSION In this study, pancreatic SWV, measured by preoperative ARFI imaging, showed significant correlations with grade of pathologic fibrosis and exocrine function of the pancreas. When pancreatic elasticity was dichotomized as “hard” or “soft,” based on the preoperative SWV, the incidence of PF after PD was significantly lower in patients with “hard” than “soft” pancreata, although the risk of postoperative PF might have also been affected by surgical procedures used in this series. These results suggest that preoperative ARFI imaging of the pancreas enables accurate estimation of the degree of fibrotic changes and exocrine function of the pancreas, both of which are associated with the risk and severity of PF after pancreatic resection. The major advantage of ARFI imaging is that it provides a quantitative measurement of pancreatic elasticity, a parameter previously evaluated subjectively by palpation of the pancreatic parenchyma during surgery. The reliability of ARFI imaging in measuring tissue elasticity was observed in patients with chronic liver diseases19,22-25 and with chronic pancreatitis.33 In patients undergoing pancreatic resection, the median (range) SWV of the pancreas was reported to be 1.51 m/s (range 0.80 to 3.40 m/s),29 which is similar to the results in our series (1.49 m/s [range 0.85 to 3.52 m/s]). In addition, the ARFI imaging method is safe, feasible, and noninvasive: using a commercially available ARFI imaging system, the SWV of the pancreas can be measured in approximately 160 seconds during preoperative work-up by conventional B-mode ultrasonography, with little inter-observer difference. A potential pitfall in ARFI imaging is that SWVs can be affected by arterial pulsation and respiratory movement of the diaphragm, as previously reported.39 When the pancreatic SWVs are
892
Harada et al
J Am Coll Surg
Preoperative Pancreatic Elasticity Assessment
Figure 4. Correlations between pancreatic shear wave velocity and (A) daily output of pancreatic juice and (B) maximum amylase concentrations in pancreatic juice after pancreatoduodenectomy. In the 35 patients who underwent pancreatoduodenectomy, (A) preoperative shear wave velocity (SWV) was negatively correlated with daily output of pancreatic juice on postoperative day (POD) 7 (r ¼ 0.667, p < 0.001). (B) Preoperative SWV was also negatively correlated with maximum amylase concentration in pancreatic juice on PODs 1 to 3 (r ¼ 0.769, p < 0.001). Open and closed circles indicate patients with “soft” (SWV <1.54 m/s, n ¼ 12) and “hard” (SWV 1.54 m/s, n ¼ 23) pancreata, respectively.
measured, ROI should be set apart from the major arteries such as the superior mesenteric artery. At present, no technique has been established to completely prevent postoperative PF. Many perioperative management strategies are strict and complicated, including those involved in closure of the pancreatic stump,40-43 pancreatic anastomosis,3,7,30,44,45 and drainage of pancreatic juice and abdominal fluid,46 all designed to prevent PF from developing into a life-threatening complication. For example, our group has used 2-stage pancreatojejunostomy for PD in patients with a thin MPD and the pancreas is soft on palpation.30 Although we found that the in-hospital mortality after PD was zero,30 this strategy has potential disadvantages, including requiring patients to wait for 3 months before secondary Table 2.
surgery, consisting of pancreatojejunostomy with an external pancreatic drainage tube. Using this technique, patients must wait until thick connective tissues cover the vessel stumps around the anastomotic site, preventing bleeding caused by PF. The ability of ARFI imaging to accurately predict the risk of postoperative PF, based on quantitative evaluation of pancreatic elasticity, may identify patients who require such meticulous surgical procedures, such as 2-stage pancreatojejunostomy, as well as those in whom even prophylactic abdominal drainage could be omitted. So, preoperative ARFI imaging may enhance the safety and efficacy of pancreatic surgery. Acoustic radiation force impulse imaging failed to measure SWVs of the pancreas in 6 of the 68 patients (9%), because massive amounts of gastrointestinal air covered
Relationship Between Shear Wave Velocity-Based Hardness of the Pancreas and Operative Outcomes
Variable
PF (ISGPF grade B or higher), n (%)* Postoperative morbidity (Clavien-Dindo grade III or higher), n (%)y Postoperative hospital stay, d, median (range)
Hard pancreas, SWV 1.54 m/s
Soft pancreas, SWV < 1.54 m/s
p Value
4 (17) 3 (10) 21 (9e65)
20 (63) 9 (28) 20 (7e77)
0.001 0.108 0.751
*PF grades A/B/C ¼ 13/20/4. y Postoperative morbidity included PF (ISGPF grade B or higher, n ¼ 24), delayed gastric emptying (n ¼ 2), cholangitis (n ¼ 1), diarrhea (n ¼ 1), and postoperative pneumonia (n ¼ 1). ISGPF, International Study Group of Pancreatic Fistula; PF, pancreatic fistula; SWV, shear wave velocity.
Vol. 219, No. 5, November 2014
Harada et al
Table 3. Multivariate Analysis of Pre- and Intraoperative Factors Predicting Postoperative Pancreatic Fistula (International Study Group of Pancreatic Fistula Grade B or Higher) Factors
Soft pancreas (SWV < 1.54 m/s) Male sex Age 70 y BMI 25.0 kg/m2 MPD diameter 3 mm Blood loss > 1,000 mL Pathology, pancreatic cancer Presence of diabetes mellitus
Multivariate analysis Odds ratio 95 % CI p Value
38.3 8.87 7.69 8.26 2.32 4.78 0.40 1.43
5.82e445 1.71e67.8 1.44e62.3 1.08e94.4 0.50e11.6 0.41e79.2 0.07e1.93 0.24e9.63
0.001 0.018 0.030 0.059 0.285 0.233 0.259 0.699
BMI, body mass index; MPD, main pancreatic duct; SWV, shear wave velocity.
the surface of the pancreas and/or patients had difficulty holding their breath during the examination. Preoperative ARFI imaging cannot be used to evaluate severely obese patients because to measure SWV, this method requires visualization of the pancreatic parenchyma within 8 cm from the surface of the body, with minimum compression to the abdominal wall. However, intraoperative ARFI imaging, done by setting a transducer directly onto the pancreatic surface, may enable measurement of pancreatic SWV even in obese patients. Another limitation of ARFI imaging is that it correlates with the degree of pathologic fibrosis but not with fatty infiltration, which may also be associated with pancreatic function and risk of postoperative PF.10,11 Additional studies are needed to enhance the ability of ARFI imaging to predict postoperative PF by, for example, including the pancreatic echogenicity reflecting fatty infiltration and intraoperative pancreatic SWVs, into estimations and optimizing SWV cut-off values in a larger series of patients.
CONCLUSIONS In conclusion, preoperative assessment of pancreatic elasticity by ARFI imaging enabled estimation of the degree of pathologic fibrosis in pancreatic tissue. Preoperative determination of pancreatic exocrine function and risk of postoperative PF may be useful in designing appropriate surgical procedures and postoperative management for individual patients undergoing pancreatic resection. Author Contributions Study conception and design: Harada, Ishizawa, Inoue Acquisition of data: Harada, Tanaka Analysis and interpretation of data: Harada, Ishizawa, Aoki, Sakamoto, Hasegawa, Sugawara
Preoperative Pancreatic Elasticity Assessment
893
Drafting of manuscript: Harada, Ishizawa, Inoue Critical revision: Fukayama, Kokudo Acknowledgment: The authors acknowledge the significant contributions made by Drs A Nakazawa, Y Okinaga, and J Kaneko, members of this study group. REFERENCES 1. Yeo CJ, Cameron JL, Sohn TA, et al. Six hundred fifty consecutive pancreaticoduodenectomies in the 1990s: pathology, complications, and outcomes. Ann Surg 1997;226:248e257. 2. Fuks D, Piessen G, Huet E, et al. Life-threatening postoperative pancreatic fistula (grade C) after pancreaticoduodenectomy: incidence, prognosis, and risk factors. Am J Surg 2009;197: 702e709. 3. Mckay A, Mackenzie S, Sutherland FR, et al. Meta-analysis of pancreaticojejunostomy versus pancreaticogastrostomy reconstruction after pancreaticoduodenectomy. Br J Surg 2006;93: 929e936. 4. Topal B, Fieuws S, Aerts R, et al. Pancreaticojejunostomy versus pancreaticogastrostomy reconstruction after pancreaticoduodenectomy for pancreatic or periampullary tumours: a multicenter randomized trial. Lancet Oncol 2013;14: 655e662. 5. Pratt WB, Maithel SK, Vanounou T, et al. Clinical and economic validation of the International Study Group of Pancreatic Fistula (ISGPF) classification scheme. Ann Surg 2007;245: 443e451. 6. Kawai M, Kondo S, Yamaue H, et al. Predictive risk factors for clinically relevant pancreatic fistula analyzed in 1,239 patients with pancreaticoduodenectomy: multicenter data collection as a project study of pancreatic surgery by the Japanese Society of Hepato-Biliary-Pancreatic Surgery. J Hepatobiliary Pancreat Sci 2011;18:601e608. 7. Yeo CJ, Cameron JL, Maher MM, et al. A prospective randomized trial of pancreaticogastrostomy versus pancreaticojejunostomy after pancreaticoduodenectomy. Ann Surg 1995; 222:588e592. 8. Yang YM, Tian XD, Zhuang Y, et al. Risk factors of pancreatic leakage after pancreaticoduodenectomy. World J Gastroenterol 2005;28:2456e2461. 9. Kah Heng CA, Salleh I, San TS, et al. Pancreatic fistula after distal pancreatectomy: incidence, risk factors and management. ANZ J Surg 2010;80:619e623. 10. Gaujoux S, Cortes A, Couvelard A, et al. Fatty pancreas and increased body mass index are risk factors of pancreatic fistula after pancreaticoduodenectomy. Surgery 2010;148:15e23. 11. Mathur A, Pitt HA, Marine M, et al. Fatty pancreas: a factor in postoperative pancreatic fistula. Ann Surg 2007;246:1058e1064. 12. Akamatsu N, Sugawara Y, Komagome M, et al. Risk factors for postoperative pancreatic fistula after pancreaticoduodenectomy: the significance of the ratio of the main pancreatic duct to the pancreas body as a predictor of leakage. J Hepatobiliary Pancreat Sci 2010;17:322e328. 13. Wada K, Traverso LW. Pancreatic anastomotic leak after the Whipple procedure is reduced using the surgical microscope. Surgery 2006;139:735e742. 14. Muscari F, Suc B, Kirzin S, et al. Risk factors for mortality and intra-abdominal complications after pancreatoduodenectomy:
894
15. 16.
17. 18. 19.
20.
21.
22.
23.
24.
25.
26. 27.
28.
29.
Harada et al
Preoperative Pancreatic Elasticity Assessment
multivariate analysis in 300 patients. Surgery 2006;139: 591e598. Pratt WB, Callery MP, Vollmer CM Jr. Risk prediction for development of pancreatic fistula using the ISGPF classification scheme. World J Surg 2008;32:419e428. Dong X, Zhang B, Kang MX, et al. Analysis of pancreatic fistula according to the International Study Group on Pancreatic Fistula classification scheme for 294 patients who underwent pancreaticoduodenectomy in a single center. Pancreas 2011; 40:222e228. Yeh TS, Jan YY, Jeng LB, et al. Pancreaticojejunal anastomotic leak after pancreaticoduodenectomyemultivariate analysis of perioperative risk factors. J Surg Res 1997;67:119e125. Piscaqlia F, Marinelli S, Bota S, et al. The role of ultrasound elastographic techniques in clinic liver disease: Current status and future perspectives. Eur J Radiol 2014;83:450e455. Friedrich-Rust M, Wunder K, Kriener S, et al. Liver fibrosis in viral hepatitis: noninvasive assessment with acoustic radiation force impulse imaging versus transient elastography. Radiology 2009;252:595e604. Bamber J, Cosqrove D, Dietrich CF, et al. EFSUMB guidelines and recommendations on the clinical use of ultrasound elastography. Part 1: Basic principles and technology. Ultraschall Med 2013;34:169e184. Fahey BJ, Nightingale KR, Nelson RC, et al. Acoustic radiation force impulse imaging of the abdomen: demonstration of feasibility and utility. Ultasound Med Biol 2005;31: 1185e1198. Guzma´n-Aroca F, Frutos-Bernal MD, Bas A, Luja´nMompea´n JA, et al. Detection of non-alcoholic steatohepatitis in patients with morbid obesity before bariatric surgery: preliminary evaluation with acoustic radiation force impulse imaging. Eur Radiol 2012;22:2525e2532. Vermehren J, Polta A, Zimmermann O, et al. Comparison of acoustic radiation force impulse imaging with transient elastography for the detection of complications in patients with cirrhosis. Liver Int 2012;32:852e858. Son CY, Kim SU, Han WK, et al. Normal liver elasticity values using acoustic radiation force impulse imaging: a prospective study in healthy living liver and kidney donors. J Gastroenterol Hepatol 2012;27:130e136. Zhang P, Zhou P, Tian SM, et al. Application of acoustic radiation force impulse imaging for the evaluation of focal liver lesion elasticity. Hepatobiliary Pancreat Dis Int 2013;12: 165e170. Meng W, Zhang G, Wu C, et al. Preliminary results of acoustic radiation force impulse (ARFI) ultrasound imaging of breast lesions. Ultrasound Med Biol 2011;37:1436e1443. Friedrich-Rust M, Romenski O, Meyer G, et al. Acoustic radiation force impulse-imaging for the evaluation of the thyroid gland: a limited patient feasibility study. Ultrasonics 2012; 52:69e74. D’Onofrio M, Crosara S, Canestrini S, et al. Virtual analysis of pancreatic cystic lesion fluid content by ultrasound acoustic radiation force impulse quantification. J Ultrasound Med 2013; 32:647e651. Lee TK, Kang CM, Park MS, et al. Prediction of postoperative pancreatic fistulas after pancreatectomy: assessment with
30. 31. 32.
33. 34. 35. 36. 37. 38.
39.
40. 41. 42.
43.
44. 45.
46.
J Am Coll Surg
acoustic radiation force impulse elastography. J Ultrasound Med 2014;33:781e786. Hasegawa K, Kokudo N, Sano K, et al. Two-stage pancreatojejunostomy in pancreaticoduodenectomy: a retrospective analysis of short-term results. Am J Surg 2008;196:3e10. Yamashita S, Sakabe M, Ishizawa T, et al. Visualization of the leakage of pancreatic juice using a chymotrypsin-activated fluorescent probe. Br J Surg 2013;100:1220e1228. Konishi T, Hiraishi M, Kubota K, et al. Segmental occlusion of the pancreatic duct with prolamine to prevent fistula formation after distal pancreatectomy. Ann Surg 1995;221: 165e170. Yashima Y, Sasahira N, Isayama H, et al. Acoustic radiation force impulse elastography for noninvasive assessment of chronic pancreatitis. J Gastroenterol 2012;47:427e432. Klo¨ppel G, Maillet B. Pseudocysts in chronic pancreatitis: a morphological analysis of 57 resection specimens and 9 autopsy pancreata. Pancreas 1991;6:266e274. Youden WJ. Index for rating diagnostic tests. Cancer 1950;3: 32e35. Yoshioka R, Saiura A, Koga R, et al. Risk factors for clinical pancreatic fistula after pancreatectomy: analysis of consecutive 100 patients. World J Surg 2010;34:121e125. Sledzianowski JF, Duffas JP, Muscari F, et al. Risk factors for mortality and intra-abdominal morbidity after distal pancreatectomy. Surgery 2005;137:180e185. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240: 205e213. Toshima T, Shirabe K, Takeishi K, et al. New method for assessing liver fibrosis based on acoustic radiation force impulse: a special reference to the difference between right and left liver. J Gastroenterol 2011;46:705e711. Mishra PK, Saluja SS, Gupta M, et al. Blumgart’s technique of pancreaticojejunostomy: an appraisal. Dig Surg 2011;28:281e287. Kennedy EP, Yeo CJ. Dunking pancreaticojejunostomy versus duct-to-mucosa anastomosis. J Hepatobiliary Pancreat Surg 2011;18:769e774. Hamilton NA, Porembka MR, Johnston FM, et al. Mesh reinforcement of pancreatic transection decreases incidence of pancreatic occlusion failure for left pancreatectomy: a single-blinded, randomized controlled trial. Ann Surg 2012;255:1037e1042. Diener MK, Seiler CM, Rossion I, et al. Efficacy of stapler versus hand-sewn closure after distal pancreatectomy (DISPACT): a randomized, controlled multicenter trial. Lancet 2011;377:1514e1522. Peng SY, Wang JW, Lau WY, et al. Conventional versus binding pancreaticojejunostomy after pancreaticoduodenectomy: a prospective randomized trial. Ann Surg 2007;245:692e698. Ke S, Ding XM, Gao J, et al. A prospective, randomized trial of Roux-en-Y reconstruction with isolated pancreatic drainage versus conventional loop reconstruction after pancreaticoduodenectomy. Surgery 2013;153:743e752. Poon RT, Fan ST, Lo CM, et al. External drainage of pancreatic duct with a stent to reduce leakage rate of pancreaticojejunostomy after pancreaticoduodenectomy: a prospective randomized trial. Ann Surg 2007;246:425e435.
Vol. 219, No. 5, November 2014
Harada et al
Preoperative Pancreatic Elasticity Assessment
894.e1
Supplementary Table 1. Reproducibility of Shear Wave Velocity Measurement of the Spleen in Healthy Volunteers HV1 Examiner
①
Spleen ②
Measured value
2.50 2.44 2.42 2.09 2.00 2.32 2.01 2.44 1.80 2.50 2.37 (1.80e2.50) 2.25 0.25
2.04 2.30 2.16 2.04 2.60 2.23 2.15 2.26 2.20 2.02 2.18 (2.02e2.60) 2.20 0.17
2.30 1.89 2.06 2.15 2.16 2.06 2.06 2.23 2.07 2.43 2.11 (1.89e2.43) 2.14 0.15 0.549
2.02 2.05 1.98 2.36 2.14 2.40 2.41 2.01 2.24 2.54 2.19 (1.98e2.54) 2.22 0.20
2.39 2.13 2.51 1.89 2.21 2.21 2.11 2.20 2.10 2.19 2.20 (1.89e2.51) 2.22 0.17
2.35 2.56 2.44 1.85 2.08 2.15 2.57 2.54 2.22 1.89 2.29 (1.85e2.57) 2.27 0.27 0.697
2.11 1.90 2.47 2.83 2.38 2.89 2.66 2.83 2.96 2.34 2.57 (1.90e2.96) 2.54 0.36
2.34 2.51 2.51 2.21 2.54 2.53 2.87 2.37 2.25 2.53 2.51 (2.21e2.87) 2.47 0.19
2.87 2.57 2.30 2.30 2.71 2.56 2.32 2.17 2.36 2.24 2.34 (2.17e2.87) 2.44 0.23 0.633
Median value Average value p value
③
HV2 Examiner
Measured value
Median value Average value p value HV3 Examiner
Measured value
Median value Average value p value
894.e2
Harada et al
J Am Coll Surg
Preoperative Pancreatic Elasticity Assessment
Supplementary Table 2. Reproducibility of Shear Wave Velocity Measurement of the Right Liver in Healthy Volunteers HV1 Examiner
①
Right Liver ②
Measured value
1.15 1.31 1.19 1.37 1.20 1.34 1.20 1.23 1.25 1.30 1.24 (1.15e1.37) 1.25 0.07
1.20 1.27 1.34 1.34 1.32 1.28 1.32 1.23 1.29 1.27 1.29 (1.20e1.34) 1.29 0.05
Median value Average value p value
③
1.25 1.31 1.30 1.29 1.31 1.30 1.28 1.34 1.27 1.37 1.3 (1.25e1.37) 1.30 0.03 0.281
HV2 Examiner
Measured value
Median value Average value p value
1.01 0.99 1.13 1.04 1.05 1.11 1.09 1.08 1.12 1.13 1.09 (0.99e1.13) 1.08 0.05
1.08 1.00 1.15 1.11 1.08 1.10 1.07 1.11 0.92 0.95 1.08 (0.92e1.15) 1.06 0.08
1.05 1.01 1.02 1.03 1.02 0.93 1.03 1.13 1.09 0.99 1.03 (0.93e1.13) 1.03 0.05 0.243
1.25 1.20 1.24 1.18 1.19 1.16 1.20 1.26 1.25 1.26 1.22 (1.16e1.26) 1.22 0.04
1.26 1.22 1.16 1.22 1.20 1.27 1.18 1.19 1.16 1.23 1.21 (1.16e1.27) 1.21 0.04
1.16 1.24 1.22 1.32 1.15 1.40 1.26 1.20 1.34 1.40 1.25 (1.15e1.40) 1.27 0.09 0.317
HV3 Examiner
Measured value
Median value Average value p value
Vol. 219, No. 5, November 2014
Harada et al
Preoperative Pancreatic Elasticity Assessment
894.e3
Supplementary Table 3. Reproducibility of Shear Wave Velocity Measurement of the Left Liver in Healthy Volunteers HV1 Examiner
①
Left Liver ②
Measured value
1.17 1.04 1.22 0.95 1.02 1.21 1.06 1.32 1.33 1.33 1.19 (0.95e1.33) 1.17 0.14
1.27 1.27 1.09 1.28 1.16 1.26 0.96 1.22 0.97 1.13 1.19 (0.96e1.28) 1.16 0.12
1.10 1.05 1.03 1.27 1.17 1.21 1.20 1.08 1.18 1.21 1.18 (1.03e1.27) 1.15 0.08 0.859
0.93 1.21 0.99 1.03 0.97 1.3 1.24 0.76 1.04 1.16 1.04 (0.76e1.30) 1.06 0.16
1.02 1.01 1.14 1.1 1.19 0.96 1.02 1.07 1.05 0.93 1.04 (0.93e1.19) 1.05 0.08
1.03 0.98 1.07 1.12 1.04 1.14 1.08 1.09 1.03 1 1.06 (0.98e1.14) 1.06 0.05 0.791
1.46 1.49 1.27 1.56 1.06 1.70 1.40 1.39 1.48 1.64 1.47 (1.06e1.70) 1.45 0.18
1.35 1.24 1.49 1.32 1.43 1.26 1.47 1.58 1.58 1.53 1.45 (1.24e1.58) 1.43 0.13
1.61 1.16 1.28 1.13 1.36 1.47 1.25 1.33 1.60 1.27 1.31 (1.13e1.61) 1.35 0.17 0.359
Median value Average value p value
③
HV2 Examiner
Measured value
Median value Average value p value HV3 Examiner
Measured value
Median value Average value p value
894.e4
Harada et al
J Am Coll Surg
Preoperative Pancreatic Elasticity Assessment
Supplementary Table 4. Reproducibility of Shear Wave Velocity Measurement of the Pancreas in Healthy Volunteers HV1 Examiner
①
Pancreas ②
Measured value
0.83 1.30 1.10 0.94 1.03 1.07 1.06 0.84 0.80 0.88 0.99 (0.80e1.30) 0.99 0.16
0.96 0.96 1.23 0.93 1.05 1.10 1.02 1.13 1.11 1.00 1.04 (0.93e1.23) 1.05 0.09
1.00 0.93 1.07 1.21 1.07 1.02 1.12 0.95 0.99 1.00 1.01 (0.93e1.21) 1.04 0.08 0.437
1.06 0.93 1.20 1.48 1.17 1.47 1.51 1.51 1.21 1.05 1.21 (0.93e1.51) 1.26 0.22
1.25 1.35 1.27 1.31 1.26 1.27 1.21 1.38 1.15 1.21 1.27 (1.15e1.38) 1.27 0.07
1.38 1.28 1.40 1.17 1.24 1.38 1.47 1.20 1.29 1.20 1.29 (1.17e1.47) 1.30 0.10 0.785
1.19 0.89 0.98 0.96 1.15 0.95 0.97 1.02 1.24 0.93 0.98 (0.89e1.24) 1.03 0.12
0.98 0.94 1.10 0.88 1.13 0.93 1.03 1.11 1.14 1.19 1.07 (0.88e1.19) 1.04 0.11
0.98 1.05 0.98 1.02 1.10 1.03 0.91 1.11 1.09 1.10 1.04 (0.91e1.11) 1.04 0.07 0.945
Median value Average value p value
③
HV2 Examiner
Measured value
Median value Average value p value HV3 Examiner
Measured value
Median value Average value p value
Vol. 219, No. 5, November 2014
Supplementary Table 5. Pancreatoduodenectomy Patient 1 Examiner
Measured value
Median value Average value p value Patient 2 Examiner
Measured value
Median value Average value p value Patient 3 Examiner
Measured value
Median value Average value p value
Harada et al
Preoperative Pancreatic Elasticity Assessment
894.e5
Reproducibility of Shear Wave Velocity Measurement of the Pancreas in Patients Undergoing
①
IPMN ②
1.51 1.72 1.34 1.70 1.66 1.39 1.62 1.32 1.66 1.38 1.57 (1.32e1.72) 1.53 0.16
1.47 1.72 1.55 1.51 1.52 1.43 1.62 1.40 1.63 1.61 1.54 (1.40e1.72) 1.55 0.10
①
Bile Duct Cancer ②
1.30 1.39 1.19 1.33 1.27 1.54 1.33 1.57 1.39 1.38 1.36 (1.19e1.57) 1.37 0.12
1.40 1.47 1.28 1.37 1.48 1.32 1.30 1.46 1.55 1.29 1.39 (1.28e1.55) 1.39 0.09
①
Pancreatic Cancer ②
0.84 0.91 1.07 0.94 0.90 1.25 1.21 1.30 0.92 1.16 1.01 (0.84e1.30) 1.05 0.17
1.17 0.99 0.92 1.09 1.21 0.99 1.00 1.18 1.22 0.95 1.05 (0.92e1.22) 1.07 0.12
③
1.69 1.27 1.53 1.75 1.37 1.68 1.45 1.49 1.66 1.67 1.60 (1.27e1.75) 1.56 0.16 0.919 ③
1.38 1.30 1.69 1.41 1.16 1.13 1.26 1.55 1.27 1.46 1.34 (1.13e1.69) 1.36 0.17 0.863 ③
0.96 1.12 0.89 1.21 0.98 0.89 1.25 0.95 1.34 0.89 0.97 (0.89e1.34) 1.05 0.17 0.927
BACKGROUND:
as severe infections and massive hemorrhage, it is important to accurately estimate the risk of PF in individual patients undergoing pancreatic resection. Several clinical factors were found to be significant predictors of postoperative PF, including soft pancreatic texture,2,6-11 small main pancreatic duct,11-15 tumor pathology other than adenocarcinoma,15,16 and larger amounts of intraoperative blood loss.6,15,17 Despite its subjectivity, hardness of the pancreas by palpation during surgery is regarded as a simple and reliable indicator for the development of postoperative PF. Quantitative methods are needed to assess the hardness of pancreatic tissue and to determine whether hardness correlates with the degree of pathologic fibrosis. Several ultrasound elastographic techniques were recently developed for noninvasive evaluation of tissue elasticity.18,19 One method, acoustic radiation force impulse (ARFI) imaging, measures the velocity of shear waves generated by tissue after irradiation with
Despite recent advances in surgical techniques and perioperative management of pancreatic resection, postoperative pancreatic fistula (PF) remains a major postoperative concern, with an incidence of 10% to 50%.1-5 Because postoperative PF can lead to life-threatening events such Disclosure Information: Nothing to disclose. Support: This work was supported by grants from the Takeda Science Foundation; the Kanae Foundation for the Promotion of Medical Science; the Ministry of Education, Culture, Sports, Science and Technology of Japan; and the Ministry of Health, Labor and Welfare of Japan. Received May 29, 2014; Revised July 17, 2014; Accepted July 30, 2014. From the Hepato-Biliary-Pancreatic Surgery Division, Department of Surgery (Harada, Ishizawa, Inoue, Aoki, Sakamoto, Hasegawa, Sugawara, Kokudo), and the Department of Pathology (Tanaka, Fukayama), Graduate School of Medicine, The University of Tokyo, Tokyo, Japan. Correspondence address: Norihiro Kokudo, MD, PhD, Hepato-BiliaryPancreatic Surgery Division, Department of Surgery, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan. email: [email protected]
ª 2014 by the American College of Surgeons Published by Elsevier Inc.
887
http://dx.doi.org/10.1016/j.jamcollsurg.2014.07.940 ISSN 1072-7515/14
888
Harada et al
Preoperative Pancreatic Elasticity Assessment
Abbreviations and Acronyms
ARFI DP MPD OR PD PF POD ROC ROI SWV
¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼
acoustic radiation force impulse distal pancreatectomy main pancreatic duct odds ratio pancreatoduodenectomy pancreatic fistula postoperative day receiver operating characteristic region of interest shear wave velocity
short-duration acoustic pulses and enables evaluation of tissue elasticity in a small (<1.0 cm) region of interest (ROI) on conventional B-mode ultrasonographic images.20,21 Using this technique, shear wave velocity (SWV) correlates with tissue hardness. In clinical settings, ARFI imaging has been used mainly to estimate the degree of hepatic fibrosis without biopsy in patients with chronic liver disease,19,22-24 and has been used for the differential diagnosis of nodular lesions in the liver,25 breast,26 thyroid,27 and pancreas.28 To our knowledge, however, ARFI imaging has not been used to estimate pancreatic hardness or whether the latter predicts a risk of PF after pancreatic surgery, except in the recent preliminary study by Lee and colleagues.29 The aim of this study was to evaluate whether the SWV of the pancreas, as measured by preoperative ARFI imaging, correlates with the pathologic degree of pancreatic fibrosis and pancreatic exocrine function in patients undergoing pancreatic resection. The ability of “soft” vs “hard” pancreata, based on preoperative SWV, to predict postoperative PF was also assessed.
J Am Coll Surg
METHODS This study was conducted with the approval of the Institutional Ethics Review Board of our institution. All patients provided written informed consent. Patients The study cohort consisted of 68 patients who underwent pancreatic resection at the University of Tokyo Hospital between February 2012 and February 2013. Preoperative ARFI imaging could not be performed in 6 of these patients because of difficulties obtaining clear and static images of the pancreatic parenchyma. In the remaining 62 patients, the SWV of the pancreas was measured by ARFI imaging before pancreatoduodenectomy (PD, n ¼ 35) or distal pancreatectomy (DP, n ¼ 27), with these SWVs compared with postoperative pancreatic exocrine function and the incidence of PF. Pancreatoduodenectomy was performed as previously described.30,31 Briefly, the pancreatic parenchyma was divided using the clamp crushing method, with meticulous ligation of small vessels. Two Roux limbs were created and used for digestive reconstruction: one for biliary and pancreatic anastomoses, and the other for gastrojejunostomy. Patients with a narrow main pancreatic duct (MPD) and/or a soft pancreatic texture, as assessed by manual palpation, underwent PD without pancreatojejunostomy, ie, totally external tube pancreatostomy, followed 3 months later by 2stage pancreatojejunostomy.30 Other patients underwent pancreatojejunostomy with duct-to-mucosa anastomosis and external tube pancreatostomy during PD. Distal pancreatectomy was also performed by dividing the pancreatic parenchyma using the clamp crushing
Figure 1. Preoperative acoustic radiation force impulse imaging of the pancreas and pathologic images of the resected specimen. (A) The shear wave velocity (SWV) of the pancreas was measured by setting a region of interest (ROI) in the pancreatic body above the confluence of the splenic and superior mesenteric veins (yellow arrows), away from the pancreatic tumor (arrowheads in left) and the superior mesenteric artery (yellow arrowhead in right). (B) Pathologic pancreatic fibrosis grading according to Klo¨ppel’s criteria. The SWVs were 1.01 m/s for normal fibrosis, 1.35 m/s for mild fibrosis, 1.70 m/s for moderate fibrosis, and 1.81 m/s for severe fibrosis.
Vol. 219, No. 5, November 2014
Harada et al
Table 1. Relationships Between Pancreatic Shear Wave Velocity and Patient Demographic and Clinical Characteristics Variables
Age, y >70 70 Sex Male Female MPD diameter, mm >3 3 Body mass index, kg/m2 >25.0 25.0 Diabetes mellitus Present Absent Pathologic diagnosis Pancreatic cancer Other diseases*
n
SWV, m/s
p Value
30 32
1.77 (1.01e3.52) 1.45 (0.85e2.37)
0.041
38 24
1.68 (0.94e3.52) 1.49 (0.85e2.50)
0.435
34 28
1.78 (1.14e3.52) 1.40 (0.85e3.19)
0.001
50 12
1.61 (0.85e3.52) 1.59 (0.94e2.37)
0.908
28 34
1.68 (0.95e3.52) 1.55 (0.85e2.98)
0.392
31 31
1.71 (0.94e3.52) 1.50 (0.85e2.98)
0.078
Continuous data are shown as median (range). *Including intraductal papillary mucinous neoplasm (n ¼ 9); carcinoma of the papilla of Vater (n ¼ 5), the lower bile duct (n ¼ 4), and the duodenum (n ¼ 2); pancreatic neuroendocrine tumor (n ¼ 5), gastric cancer (n ¼ 2), serous cystic neoplasm (n ¼ 2), solid-pseudopapillary neoplasm (n ¼ 1), and chronic pancreatitis (n ¼ 1). MPD, main pancreatic duct; SWV, shear wave velocity.
method and by ligating the small vessels and the MPD.32 Prophylactic abdominal drains were routinely placed on the superior and inferior aspects of the pancreatic stump. Preoperative acoustic radiation force impulse imaging of the pancreas Preoperative ARFI imaging was performed using the ACUSON S2000 (Siemens Medical Solutions) in the Virtual Touch Tissue Quantification mode, as described.19-29,33 Briefly, patients in the fasting state were placed in the decubitus position. To accurately measure SWVs, conventional B-mode ultrasonography was performed to visualize the pancreatic parenchyma within a depth of 8 cm from the transducer.21,33 An ROI measuring 10 mm axially by 6 mm in width was set in the pancreatic body above the confluence of the splenic and portal veins (splenoportal junction), away from the pancreatic tumors and the superior mesenteric artery (Fig. 1A). Before using preoperative ARFI imaging to evaluate the study population, the reproducibility of this technique was validated in 3 healthy volunteers (3 men, ages 24, 31, and 38 years) and in 3 patients (2 men
Preoperative Pancreatic Elasticity Assessment
889
and 1 woman, ages 71, 72, and 82 years) who underwent PD for biliary or pancreatic malignancies. Three surgeons, blinded to results obtained by the others, measured SWVs 10 times in each subject. In this validation trial, pancreatic SWVs measured by the 3 surgeons did not differ significantly, either in the group of healthy volunteers or in the group of patients undergoing PD (ANOVA, Appendix 1: Supplementary Tables 1 to 5, online only). Shear wave velocities in the study population were measured 5 times in each of the 62 patients; the mean SWV of the median 3 values for each patient was calculated and used in further analyses. Pathologic assessment of the pancreas Pancreatic tissue samples were obtained 0.5 to 1.0 cm from the pancreatic stump, fixed in formalin, and stained with hematoxylin and eosin. These slides were evaluated by an experienced pathologist (MT), blinded to each patient’s clinical data and postoperative course. The degree of pancreatic fibrosis was scored according to perilobular and intralobular fibrosis and divided into 4 categories of Klo¨ppel scores: 1 to 3, normal tissue; 4 to 6, mild fibrosis; 7 to 9, moderate fibrosis; and 10 to 12, severe fibrosis (Fig. 1B).34 Fatty infiltration into pancreatic tissue was evaluated microscopically as 1 to 2, no fatty infiltration or 3 to 4, fatty infiltration.10 Based on the results of SWVs and pancreatic fibrosis score, cut-off values of SWV predicting pathologic pancreatic fibrosis, defined as a Klo¨ppel score 7, were calculated using a receiver operating characteristic (ROC) curve and a common optimization step maximizing Youden’s index.35 Evaluation of pancreatic exocrine function and postoperative pancreatic fistula As indicators of pancreatic exocrine function, amylase concentrations and daily amounts of pancreatic juice were evaluated using fluid samples drained from an external pancreatic tube in 35 patients who underwent PD. The amylase concentrations in pancreatic juice were not evaluated in 5 patients; 2 showed no discharge of pancreatic juice after surgery, and amylase concentrations were not determined in the other 3. Postoperative PF was defined as an amylase concentration in fluid more than 3 times the upper limit of serum amylase concentration after the third postoperative day.5 Statistical analyses Continuous data were expressed as median (range). Quantitative and categorized variables were compared using the Wilcoxon rank-sum test and Fisher’s exact test, respectively. The correlations of SWV with fibrosis score, fatty
890
Harada et al
Preoperative Pancreatic Elasticity Assessment
J Am Coll Surg
Figure 2. Correlations between pancreatic shear wave velocity and pathologic findings of resected specimens. The preoperative shear wave velocity (SWV) of the pancreas was positively correlated with (A) pathologic fibrosis score (Spearman’s rank correlation coefficient [r] ¼ 0.660, p < 0.001), but not with (B) pathologic fatty infiltration score (r ¼ 0.058, p ¼ 0.654).
infiltration score, and pancreatic exocrine function were evaluated using Spearman’s rank correlation test. Multivariate analysis was performed to evaluate the ability of preoperative SWV to predict the risk of postoperative PF. Due to their clinical significance, 8 potentially important preoperative and intraoperative factors were included in the analysis: age (70 years vs >70 years),36 sex,6 diabetes mellitus (yes vs no), body mass index (25 kg/m2 vs >25 kg/m2),6,37 pathology of the tumor (pancreatic cancer vs other neoplasm),15,16 amount of blood loss (1,000 mL vs >1,000 mL),6,15,17 MPD diameter (3 mm vs >3 mm),12,13 and SWV (1.54 m/s vs >1.54 m/s, based on ROC curve analysis). The results were expressed as adjusted odds ratios (OR) with 95% confidence intervals (CI), and p values were calculated using the likelihood ratio test. Values of p < 0.05 were considered statistically significant. All statistical analyses were performed using JMP software version 9.0.0 (SAS Institute).
RESULTS Shear wave velocities of the pancreas and patients’ background characteristics The study subjects consisted of 38 men and 24 women, ranging in age from 28 to 85 years (median 70 years).
Median preoperative pancreatic SWV was 1.49 m/s (range 0.85 to 3.52 m/s). Preoperative SWVs of the pancreas were higher in patients aged 70 years or younger and in patients with an MPD diameter of 3 mm or less (Table 1). The median time to measure SWV of the pancreas was 157 seconds (range 42 to 490 seconds). Shear wave velocities of the pancreas and pathologic diagnosis of fibrosis and fatty infiltration Preoperative SWV of the pancreas significantly correlated with pathologic fibrosis score (Spearman’s rank correlation coefficient [r] ¼ 0.660, p < 0.001; Fig. 2A), but not with pathologic fatty infiltration score (r ¼ 0.058, p ¼ 0.654; Fig. 2B). Receiver operating characteristic curve analysis was performed to determine a cut-off value of SWV predictive of pancreatic fibrosis, defined as a Klo¨ppel score 7 (Fig. 3). An SWV cut-off of 1.54 m/s yielded an area under the curve of 0.881 (p < 0.001). The sensitivity, specificity, positive predictive value, and negative predictive value of this cut-off were 91%, 75%, 67%, and 93%, respectively. Based on these results, the 62 patients enrolled in this study were divided into soft (SWVs < 1.54 m/s, n ¼ 32) and hard (SWV 1.54 m/s, n ¼ 30) pancreas groups in the subsequent analyses.
Vol. 219, No. 5, November 2014
Harada et al
Preoperative Pancreatic Elasticity Assessment
891
and 71% (5 of 7) in the patients who had undergone primary and secondary pancreatojejunostomy. The incidence of postoperative PF (63% vs 17%, p ¼ 0.001) and the morbidity rate, defined as Clavien-Dindo grade III or higher38 (28% vs 10%, p ¼ 0.108) were significantly higher in the soft than in the hard pancreas group (Table 2). Multivariate analysis revealed that soft pancreas, defined as a preoperative SWV < 1.54 m/s, was an independent predictive factor of postoperative PF (odd ratio [OR] 38.3, 95% CI 5.82 to 445; p ¼ 0.001). Other independent predictors of postoperative PF were age 70 years (OR 7.69, 95% CI 1.44 to 62.3; p ¼ 0.030) and male sex (OR 8.87, 95% CI 1.71 to 67.8; p ¼ 0.018, Table 3).
Figure 3. Receiver operating characteristic curves analysis of shear wave velocity for predicting pathologic fibrosis of the pancreas. Receiver operating characteristic curve analysis was used to determine the cut-off value of shear wave velocity (SWV) for predicting pancreatic fibrosis, defined as a Klo¨ppel score 7. When the SWV cut-off was set at 1.54 m/s, it provided an area under the curve of 0.881 (p < 0.001) with sensitivity, specificity, positive predictive value, and negative predictive value of 91%, 75%, 67%, and 93%, respectively.
Shear wave velocities of the pancreas and pancreatic exocrine function In the 35 patients who underwent PD, higher preoperative SWV of the pancreas was significantly correlated with a lower volume of pancreatic juice on postoperative day (POD) 7 (Fig. 4A) and lower maximum amylase concentrations in pancreatic juice on PODs 1 and 3 (Fig. 4B). Of the 35 patients who underwent PD, 23 had a hard and 12 had a soft pancreas, with the former having significantly lower median volumes of pancreatic juice on POD 7 (40 mL [range 0 to 155 mL] vs (215 mL [range 130 to 420 mL], p < 0.001) and significantly lower median maximum amylase concentrations in pancreatic juice on PODs 1 to 3 (3,062 units/L [range 2 to 16,350 units/L]) vs 22,747 units/L [range 8,154 to 54,270 units/L], p < 0.001). Shear wave velocity-based hardness of the pancreas and operative outcomes Postoperative PF (International Study Group on Pancreatic Fistula [ISGPF] grade B or higher) developed in 12 of the 35 patients who underwent PD (34%) and 27 of those who underwent DP (44%). Among the 35 PD patients, the incidences of postoperative PF were 25% (7 of 28)
DISCUSSION In this study, pancreatic SWV, measured by preoperative ARFI imaging, showed significant correlations with grade of pathologic fibrosis and exocrine function of the pancreas. When pancreatic elasticity was dichotomized as “hard” or “soft,” based on the preoperative SWV, the incidence of PF after PD was significantly lower in patients with “hard” than “soft” pancreata, although the risk of postoperative PF might have also been affected by surgical procedures used in this series. These results suggest that preoperative ARFI imaging of the pancreas enables accurate estimation of the degree of fibrotic changes and exocrine function of the pancreas, both of which are associated with the risk and severity of PF after pancreatic resection. The major advantage of ARFI imaging is that it provides a quantitative measurement of pancreatic elasticity, a parameter previously evaluated subjectively by palpation of the pancreatic parenchyma during surgery. The reliability of ARFI imaging in measuring tissue elasticity was observed in patients with chronic liver diseases19,22-25 and with chronic pancreatitis.33 In patients undergoing pancreatic resection, the median (range) SWV of the pancreas was reported to be 1.51 m/s (range 0.80 to 3.40 m/s),29 which is similar to the results in our series (1.49 m/s [range 0.85 to 3.52 m/s]). In addition, the ARFI imaging method is safe, feasible, and noninvasive: using a commercially available ARFI imaging system, the SWV of the pancreas can be measured in approximately 160 seconds during preoperative work-up by conventional B-mode ultrasonography, with little inter-observer difference. A potential pitfall in ARFI imaging is that SWVs can be affected by arterial pulsation and respiratory movement of the diaphragm, as previously reported.39 When the pancreatic SWVs are
892
Harada et al
J Am Coll Surg
Preoperative Pancreatic Elasticity Assessment
Figure 4. Correlations between pancreatic shear wave velocity and (A) daily output of pancreatic juice and (B) maximum amylase concentrations in pancreatic juice after pancreatoduodenectomy. In the 35 patients who underwent pancreatoduodenectomy, (A) preoperative shear wave velocity (SWV) was negatively correlated with daily output of pancreatic juice on postoperative day (POD) 7 (r ¼ 0.667, p < 0.001). (B) Preoperative SWV was also negatively correlated with maximum amylase concentration in pancreatic juice on PODs 1 to 3 (r ¼ 0.769, p < 0.001). Open and closed circles indicate patients with “soft” (SWV <1.54 m/s, n ¼ 12) and “hard” (SWV 1.54 m/s, n ¼ 23) pancreata, respectively.
measured, ROI should be set apart from the major arteries such as the superior mesenteric artery. At present, no technique has been established to completely prevent postoperative PF. Many perioperative management strategies are strict and complicated, including those involved in closure of the pancreatic stump,40-43 pancreatic anastomosis,3,7,30,44,45 and drainage of pancreatic juice and abdominal fluid,46 all designed to prevent PF from developing into a life-threatening complication. For example, our group has used 2-stage pancreatojejunostomy for PD in patients with a thin MPD and the pancreas is soft on palpation.30 Although we found that the in-hospital mortality after PD was zero,30 this strategy has potential disadvantages, including requiring patients to wait for 3 months before secondary Table 2.
surgery, consisting of pancreatojejunostomy with an external pancreatic drainage tube. Using this technique, patients must wait until thick connective tissues cover the vessel stumps around the anastomotic site, preventing bleeding caused by PF. The ability of ARFI imaging to accurately predict the risk of postoperative PF, based on quantitative evaluation of pancreatic elasticity, may identify patients who require such meticulous surgical procedures, such as 2-stage pancreatojejunostomy, as well as those in whom even prophylactic abdominal drainage could be omitted. So, preoperative ARFI imaging may enhance the safety and efficacy of pancreatic surgery. Acoustic radiation force impulse imaging failed to measure SWVs of the pancreas in 6 of the 68 patients (9%), because massive amounts of gastrointestinal air covered
Relationship Between Shear Wave Velocity-Based Hardness of the Pancreas and Operative Outcomes
Variable
PF (ISGPF grade B or higher), n (%)* Postoperative morbidity (Clavien-Dindo grade III or higher), n (%)y Postoperative hospital stay, d, median (range)
Hard pancreas, SWV 1.54 m/s
Soft pancreas, SWV < 1.54 m/s
p Value
4 (17) 3 (10) 21 (9e65)
20 (63) 9 (28) 20 (7e77)
0.001 0.108 0.751
*PF grades A/B/C ¼ 13/20/4. y Postoperative morbidity included PF (ISGPF grade B or higher, n ¼ 24), delayed gastric emptying (n ¼ 2), cholangitis (n ¼ 1), diarrhea (n ¼ 1), and postoperative pneumonia (n ¼ 1). ISGPF, International Study Group of Pancreatic Fistula; PF, pancreatic fistula; SWV, shear wave velocity.
Vol. 219, No. 5, November 2014
Harada et al
Table 3. Multivariate Analysis of Pre- and Intraoperative Factors Predicting Postoperative Pancreatic Fistula (International Study Group of Pancreatic Fistula Grade B or Higher) Factors
Soft pancreas (SWV < 1.54 m/s) Male sex Age 70 y BMI 25.0 kg/m2 MPD diameter 3 mm Blood loss > 1,000 mL Pathology, pancreatic cancer Presence of diabetes mellitus
Multivariate analysis Odds ratio 95 % CI p Value
38.3 8.87 7.69 8.26 2.32 4.78 0.40 1.43
5.82e445 1.71e67.8 1.44e62.3 1.08e94.4 0.50e11.6 0.41e79.2 0.07e1.93 0.24e9.63
0.001 0.018 0.030 0.059 0.285 0.233 0.259 0.699
BMI, body mass index; MPD, main pancreatic duct; SWV, shear wave velocity.
the surface of the pancreas and/or patients had difficulty holding their breath during the examination. Preoperative ARFI imaging cannot be used to evaluate severely obese patients because to measure SWV, this method requires visualization of the pancreatic parenchyma within 8 cm from the surface of the body, with minimum compression to the abdominal wall. However, intraoperative ARFI imaging, done by setting a transducer directly onto the pancreatic surface, may enable measurement of pancreatic SWV even in obese patients. Another limitation of ARFI imaging is that it correlates with the degree of pathologic fibrosis but not with fatty infiltration, which may also be associated with pancreatic function and risk of postoperative PF.10,11 Additional studies are needed to enhance the ability of ARFI imaging to predict postoperative PF by, for example, including the pancreatic echogenicity reflecting fatty infiltration and intraoperative pancreatic SWVs, into estimations and optimizing SWV cut-off values in a larger series of patients.
CONCLUSIONS In conclusion, preoperative assessment of pancreatic elasticity by ARFI imaging enabled estimation of the degree of pathologic fibrosis in pancreatic tissue. Preoperative determination of pancreatic exocrine function and risk of postoperative PF may be useful in designing appropriate surgical procedures and postoperative management for individual patients undergoing pancreatic resection. Author Contributions Study conception and design: Harada, Ishizawa, Inoue Acquisition of data: Harada, Tanaka Analysis and interpretation of data: Harada, Ishizawa, Aoki, Sakamoto, Hasegawa, Sugawara
Preoperative Pancreatic Elasticity Assessment
893
Drafting of manuscript: Harada, Ishizawa, Inoue Critical revision: Fukayama, Kokudo Acknowledgment: The authors acknowledge the significant contributions made by Drs A Nakazawa, Y Okinaga, and J Kaneko, members of this study group. REFERENCES 1. Yeo CJ, Cameron JL, Sohn TA, et al. Six hundred fifty consecutive pancreaticoduodenectomies in the 1990s: pathology, complications, and outcomes. Ann Surg 1997;226:248e257. 2. Fuks D, Piessen G, Huet E, et al. Life-threatening postoperative pancreatic fistula (grade C) after pancreaticoduodenectomy: incidence, prognosis, and risk factors. Am J Surg 2009;197: 702e709. 3. Mckay A, Mackenzie S, Sutherland FR, et al. Meta-analysis of pancreaticojejunostomy versus pancreaticogastrostomy reconstruction after pancreaticoduodenectomy. Br J Surg 2006;93: 929e936. 4. Topal B, Fieuws S, Aerts R, et al. Pancreaticojejunostomy versus pancreaticogastrostomy reconstruction after pancreaticoduodenectomy for pancreatic or periampullary tumours: a multicenter randomized trial. Lancet Oncol 2013;14: 655e662. 5. Pratt WB, Maithel SK, Vanounou T, et al. Clinical and economic validation of the International Study Group of Pancreatic Fistula (ISGPF) classification scheme. Ann Surg 2007;245: 443e451. 6. Kawai M, Kondo S, Yamaue H, et al. Predictive risk factors for clinically relevant pancreatic fistula analyzed in 1,239 patients with pancreaticoduodenectomy: multicenter data collection as a project study of pancreatic surgery by the Japanese Society of Hepato-Biliary-Pancreatic Surgery. J Hepatobiliary Pancreat Sci 2011;18:601e608. 7. Yeo CJ, Cameron JL, Maher MM, et al. A prospective randomized trial of pancreaticogastrostomy versus pancreaticojejunostomy after pancreaticoduodenectomy. Ann Surg 1995; 222:588e592. 8. Yang YM, Tian XD, Zhuang Y, et al. Risk factors of pancreatic leakage after pancreaticoduodenectomy. World J Gastroenterol 2005;28:2456e2461. 9. Kah Heng CA, Salleh I, San TS, et al. Pancreatic fistula after distal pancreatectomy: incidence, risk factors and management. ANZ J Surg 2010;80:619e623. 10. Gaujoux S, Cortes A, Couvelard A, et al. Fatty pancreas and increased body mass index are risk factors of pancreatic fistula after pancreaticoduodenectomy. Surgery 2010;148:15e23. 11. Mathur A, Pitt HA, Marine M, et al. Fatty pancreas: a factor in postoperative pancreatic fistula. Ann Surg 2007;246:1058e1064. 12. Akamatsu N, Sugawara Y, Komagome M, et al. Risk factors for postoperative pancreatic fistula after pancreaticoduodenectomy: the significance of the ratio of the main pancreatic duct to the pancreas body as a predictor of leakage. J Hepatobiliary Pancreat Sci 2010;17:322e328. 13. Wada K, Traverso LW. Pancreatic anastomotic leak after the Whipple procedure is reduced using the surgical microscope. Surgery 2006;139:735e742. 14. Muscari F, Suc B, Kirzin S, et al. Risk factors for mortality and intra-abdominal complications after pancreatoduodenectomy:
894
15. 16.
17. 18. 19.
20.
21.
22.
23.
24.
25.
26. 27.
28.
29.
Harada et al
Preoperative Pancreatic Elasticity Assessment
multivariate analysis in 300 patients. Surgery 2006;139: 591e598. Pratt WB, Callery MP, Vollmer CM Jr. Risk prediction for development of pancreatic fistula using the ISGPF classification scheme. World J Surg 2008;32:419e428. Dong X, Zhang B, Kang MX, et al. Analysis of pancreatic fistula according to the International Study Group on Pancreatic Fistula classification scheme for 294 patients who underwent pancreaticoduodenectomy in a single center. Pancreas 2011; 40:222e228. Yeh TS, Jan YY, Jeng LB, et al. Pancreaticojejunal anastomotic leak after pancreaticoduodenectomyemultivariate analysis of perioperative risk factors. J Surg Res 1997;67:119e125. Piscaqlia F, Marinelli S, Bota S, et al. The role of ultrasound elastographic techniques in clinic liver disease: Current status and future perspectives. Eur J Radiol 2014;83:450e455. Friedrich-Rust M, Wunder K, Kriener S, et al. Liver fibrosis in viral hepatitis: noninvasive assessment with acoustic radiation force impulse imaging versus transient elastography. Radiology 2009;252:595e604. Bamber J, Cosqrove D, Dietrich CF, et al. EFSUMB guidelines and recommendations on the clinical use of ultrasound elastography. Part 1: Basic principles and technology. Ultraschall Med 2013;34:169e184. Fahey BJ, Nightingale KR, Nelson RC, et al. Acoustic radiation force impulse imaging of the abdomen: demonstration of feasibility and utility. Ultasound Med Biol 2005;31: 1185e1198. Guzma´n-Aroca F, Frutos-Bernal MD, Bas A, Luja´nMompea´n JA, et al. Detection of non-alcoholic steatohepatitis in patients with morbid obesity before bariatric surgery: preliminary evaluation with acoustic radiation force impulse imaging. Eur Radiol 2012;22:2525e2532. Vermehren J, Polta A, Zimmermann O, et al. Comparison of acoustic radiation force impulse imaging with transient elastography for the detection of complications in patients with cirrhosis. Liver Int 2012;32:852e858. Son CY, Kim SU, Han WK, et al. Normal liver elasticity values using acoustic radiation force impulse imaging: a prospective study in healthy living liver and kidney donors. J Gastroenterol Hepatol 2012;27:130e136. Zhang P, Zhou P, Tian SM, et al. Application of acoustic radiation force impulse imaging for the evaluation of focal liver lesion elasticity. Hepatobiliary Pancreat Dis Int 2013;12: 165e170. Meng W, Zhang G, Wu C, et al. Preliminary results of acoustic radiation force impulse (ARFI) ultrasound imaging of breast lesions. Ultrasound Med Biol 2011;37:1436e1443. Friedrich-Rust M, Romenski O, Meyer G, et al. Acoustic radiation force impulse-imaging for the evaluation of the thyroid gland: a limited patient feasibility study. Ultrasonics 2012; 52:69e74. D’Onofrio M, Crosara S, Canestrini S, et al. Virtual analysis of pancreatic cystic lesion fluid content by ultrasound acoustic radiation force impulse quantification. J Ultrasound Med 2013; 32:647e651. Lee TK, Kang CM, Park MS, et al. Prediction of postoperative pancreatic fistulas after pancreatectomy: assessment with
30. 31. 32.
33. 34. 35. 36. 37. 38.
39.
40. 41. 42.
43.
44. 45.
46.
J Am Coll Surg
acoustic radiation force impulse elastography. J Ultrasound Med 2014;33:781e786. Hasegawa K, Kokudo N, Sano K, et al. Two-stage pancreatojejunostomy in pancreaticoduodenectomy: a retrospective analysis of short-term results. Am J Surg 2008;196:3e10. Yamashita S, Sakabe M, Ishizawa T, et al. Visualization of the leakage of pancreatic juice using a chymotrypsin-activated fluorescent probe. Br J Surg 2013;100:1220e1228. Konishi T, Hiraishi M, Kubota K, et al. Segmental occlusion of the pancreatic duct with prolamine to prevent fistula formation after distal pancreatectomy. Ann Surg 1995;221: 165e170. Yashima Y, Sasahira N, Isayama H, et al. Acoustic radiation force impulse elastography for noninvasive assessment of chronic pancreatitis. J Gastroenterol 2012;47:427e432. Klo¨ppel G, Maillet B. Pseudocysts in chronic pancreatitis: a morphological analysis of 57 resection specimens and 9 autopsy pancreata. Pancreas 1991;6:266e274. Youden WJ. Index for rating diagnostic tests. Cancer 1950;3: 32e35. Yoshioka R, Saiura A, Koga R, et al. Risk factors for clinical pancreatic fistula after pancreatectomy: analysis of consecutive 100 patients. World J Surg 2010;34:121e125. Sledzianowski JF, Duffas JP, Muscari F, et al. Risk factors for mortality and intra-abdominal morbidity after distal pancreatectomy. Surgery 2005;137:180e185. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240: 205e213. Toshima T, Shirabe K, Takeishi K, et al. New method for assessing liver fibrosis based on acoustic radiation force impulse: a special reference to the difference between right and left liver. J Gastroenterol 2011;46:705e711. Mishra PK, Saluja SS, Gupta M, et al. Blumgart’s technique of pancreaticojejunostomy: an appraisal. Dig Surg 2011;28:281e287. Kennedy EP, Yeo CJ. Dunking pancreaticojejunostomy versus duct-to-mucosa anastomosis. J Hepatobiliary Pancreat Surg 2011;18:769e774. Hamilton NA, Porembka MR, Johnston FM, et al. Mesh reinforcement of pancreatic transection decreases incidence of pancreatic occlusion failure for left pancreatectomy: a single-blinded, randomized controlled trial. Ann Surg 2012;255:1037e1042. Diener MK, Seiler CM, Rossion I, et al. Efficacy of stapler versus hand-sewn closure after distal pancreatectomy (DISPACT): a randomized, controlled multicenter trial. Lancet 2011;377:1514e1522. Peng SY, Wang JW, Lau WY, et al. Conventional versus binding pancreaticojejunostomy after pancreaticoduodenectomy: a prospective randomized trial. Ann Surg 2007;245:692e698. Ke S, Ding XM, Gao J, et al. A prospective, randomized trial of Roux-en-Y reconstruction with isolated pancreatic drainage versus conventional loop reconstruction after pancreaticoduodenectomy. Surgery 2013;153:743e752. Poon RT, Fan ST, Lo CM, et al. External drainage of pancreatic duct with a stent to reduce leakage rate of pancreaticojejunostomy after pancreaticoduodenectomy: a prospective randomized trial. Ann Surg 2007;246:425e435.
Vol. 219, No. 5, November 2014
Harada et al
Preoperative Pancreatic Elasticity Assessment
894.e1
Supplementary Table 1. Reproducibility of Shear Wave Velocity Measurement of the Spleen in Healthy Volunteers HV1 Examiner
①
Spleen ②
Measured value
2.50 2.44 2.42 2.09 2.00 2.32 2.01 2.44 1.80 2.50 2.37 (1.80e2.50) 2.25 0.25
2.04 2.30 2.16 2.04 2.60 2.23 2.15 2.26 2.20 2.02 2.18 (2.02e2.60) 2.20 0.17
2.30 1.89 2.06 2.15 2.16 2.06 2.06 2.23 2.07 2.43 2.11 (1.89e2.43) 2.14 0.15 0.549
2.02 2.05 1.98 2.36 2.14 2.40 2.41 2.01 2.24 2.54 2.19 (1.98e2.54) 2.22 0.20
2.39 2.13 2.51 1.89 2.21 2.21 2.11 2.20 2.10 2.19 2.20 (1.89e2.51) 2.22 0.17
2.35 2.56 2.44 1.85 2.08 2.15 2.57 2.54 2.22 1.89 2.29 (1.85e2.57) 2.27 0.27 0.697
2.11 1.90 2.47 2.83 2.38 2.89 2.66 2.83 2.96 2.34 2.57 (1.90e2.96) 2.54 0.36
2.34 2.51 2.51 2.21 2.54 2.53 2.87 2.37 2.25 2.53 2.51 (2.21e2.87) 2.47 0.19
2.87 2.57 2.30 2.30 2.71 2.56 2.32 2.17 2.36 2.24 2.34 (2.17e2.87) 2.44 0.23 0.633
Median value Average value p value
③
HV2 Examiner
Measured value
Median value Average value p value HV3 Examiner
Measured value
Median value Average value p value
894.e2
Harada et al
J Am Coll Surg
Preoperative Pancreatic Elasticity Assessment
Supplementary Table 2. Reproducibility of Shear Wave Velocity Measurement of the Right Liver in Healthy Volunteers HV1 Examiner
①
Right Liver ②
Measured value
1.15 1.31 1.19 1.37 1.20 1.34 1.20 1.23 1.25 1.30 1.24 (1.15e1.37) 1.25 0.07
1.20 1.27 1.34 1.34 1.32 1.28 1.32 1.23 1.29 1.27 1.29 (1.20e1.34) 1.29 0.05
Median value Average value p value
③
1.25 1.31 1.30 1.29 1.31 1.30 1.28 1.34 1.27 1.37 1.3 (1.25e1.37) 1.30 0.03 0.281
HV2 Examiner
Measured value
Median value Average value p value
1.01 0.99 1.13 1.04 1.05 1.11 1.09 1.08 1.12 1.13 1.09 (0.99e1.13) 1.08 0.05
1.08 1.00 1.15 1.11 1.08 1.10 1.07 1.11 0.92 0.95 1.08 (0.92e1.15) 1.06 0.08
1.05 1.01 1.02 1.03 1.02 0.93 1.03 1.13 1.09 0.99 1.03 (0.93e1.13) 1.03 0.05 0.243
1.25 1.20 1.24 1.18 1.19 1.16 1.20 1.26 1.25 1.26 1.22 (1.16e1.26) 1.22 0.04
1.26 1.22 1.16 1.22 1.20 1.27 1.18 1.19 1.16 1.23 1.21 (1.16e1.27) 1.21 0.04
1.16 1.24 1.22 1.32 1.15 1.40 1.26 1.20 1.34 1.40 1.25 (1.15e1.40) 1.27 0.09 0.317
HV3 Examiner
Measured value
Median value Average value p value
Vol. 219, No. 5, November 2014
Harada et al
Preoperative Pancreatic Elasticity Assessment
894.e3
Supplementary Table 3. Reproducibility of Shear Wave Velocity Measurement of the Left Liver in Healthy Volunteers HV1 Examiner
①
Left Liver ②
Measured value
1.17 1.04 1.22 0.95 1.02 1.21 1.06 1.32 1.33 1.33 1.19 (0.95e1.33) 1.17 0.14
1.27 1.27 1.09 1.28 1.16 1.26 0.96 1.22 0.97 1.13 1.19 (0.96e1.28) 1.16 0.12
1.10 1.05 1.03 1.27 1.17 1.21 1.20 1.08 1.18 1.21 1.18 (1.03e1.27) 1.15 0.08 0.859
0.93 1.21 0.99 1.03 0.97 1.3 1.24 0.76 1.04 1.16 1.04 (0.76e1.30) 1.06 0.16
1.02 1.01 1.14 1.1 1.19 0.96 1.02 1.07 1.05 0.93 1.04 (0.93e1.19) 1.05 0.08
1.03 0.98 1.07 1.12 1.04 1.14 1.08 1.09 1.03 1 1.06 (0.98e1.14) 1.06 0.05 0.791
1.46 1.49 1.27 1.56 1.06 1.70 1.40 1.39 1.48 1.64 1.47 (1.06e1.70) 1.45 0.18
1.35 1.24 1.49 1.32 1.43 1.26 1.47 1.58 1.58 1.53 1.45 (1.24e1.58) 1.43 0.13
1.61 1.16 1.28 1.13 1.36 1.47 1.25 1.33 1.60 1.27 1.31 (1.13e1.61) 1.35 0.17 0.359
Median value Average value p value
③
HV2 Examiner
Measured value
Median value Average value p value HV3 Examiner
Measured value
Median value Average value p value
894.e4
Harada et al
J Am Coll Surg
Preoperative Pancreatic Elasticity Assessment
Supplementary Table 4. Reproducibility of Shear Wave Velocity Measurement of the Pancreas in Healthy Volunteers HV1 Examiner
①
Pancreas ②
Measured value
0.83 1.30 1.10 0.94 1.03 1.07 1.06 0.84 0.80 0.88 0.99 (0.80e1.30) 0.99 0.16
0.96 0.96 1.23 0.93 1.05 1.10 1.02 1.13 1.11 1.00 1.04 (0.93e1.23) 1.05 0.09
1.00 0.93 1.07 1.21 1.07 1.02 1.12 0.95 0.99 1.00 1.01 (0.93e1.21) 1.04 0.08 0.437
1.06 0.93 1.20 1.48 1.17 1.47 1.51 1.51 1.21 1.05 1.21 (0.93e1.51) 1.26 0.22
1.25 1.35 1.27 1.31 1.26 1.27 1.21 1.38 1.15 1.21 1.27 (1.15e1.38) 1.27 0.07
1.38 1.28 1.40 1.17 1.24 1.38 1.47 1.20 1.29 1.20 1.29 (1.17e1.47) 1.30 0.10 0.785
1.19 0.89 0.98 0.96 1.15 0.95 0.97 1.02 1.24 0.93 0.98 (0.89e1.24) 1.03 0.12
0.98 0.94 1.10 0.88 1.13 0.93 1.03 1.11 1.14 1.19 1.07 (0.88e1.19) 1.04 0.11
0.98 1.05 0.98 1.02 1.10 1.03 0.91 1.11 1.09 1.10 1.04 (0.91e1.11) 1.04 0.07 0.945
Median value Average value p value
③
HV2 Examiner
Measured value
Median value Average value p value HV3 Examiner
Measured value
Median value Average value p value
Vol. 219, No. 5, November 2014
Supplementary Table 5. Pancreatoduodenectomy Patient 1 Examiner
Measured value
Median value Average value p value Patient 2 Examiner
Measured value
Median value Average value p value Patient 3 Examiner
Measured value
Median value Average value p value
Harada et al
Preoperative Pancreatic Elasticity Assessment
894.e5
Reproducibility of Shear Wave Velocity Measurement of the Pancreas in Patients Undergoing
①
IPMN ②
1.51 1.72 1.34 1.70 1.66 1.39 1.62 1.32 1.66 1.38 1.57 (1.32e1.72) 1.53 0.16
1.47 1.72 1.55 1.51 1.52 1.43 1.62 1.40 1.63 1.61 1.54 (1.40e1.72) 1.55 0.10
①
Bile Duct Cancer ②
1.30 1.39 1.19 1.33 1.27 1.54 1.33 1.57 1.39 1.38 1.36 (1.19e1.57) 1.37 0.12
1.40 1.47 1.28 1.37 1.48 1.32 1.30 1.46 1.55 1.29 1.39 (1.28e1.55) 1.39 0.09
①
Pancreatic Cancer ②
0.84 0.91 1.07 0.94 0.90 1.25 1.21 1.30 0.92 1.16 1.01 (0.84e1.30) 1.05 0.17
1.17 0.99 0.92 1.09 1.21 0.99 1.00 1.18 1.22 0.95 1.05 (0.92e1.22) 1.07 0.12
③
1.69 1.27 1.53 1.75 1.37 1.68 1.45 1.49 1.66 1.67 1.60 (1.27e1.75) 1.56 0.16 0.919 ③
1.38 1.30 1.69 1.41 1.16 1.13 1.26 1.55 1.27 1.46 1.34 (1.13e1.69) 1.36 0.17 0.863 ③
0.96 1.12 0.89 1.21 0.98 0.89 1.25 0.95 1.34 0.89 0.97 (0.89e1.34) 1.05 0.17 0.927