j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 5 ) 1 e5
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.JournalofSurgicalResearch.com
The diagnostic pathway for solid pancreatic neoplasms: are we applying too many tests? Michael R. Driedger, MD,a Elijah Dixon, MD, MSC, FRCSC, FACS,a Rachid Mohamed, MD, FRCPC,b Francis R. Sutherland, MD, FRCSC,a Oliver F. Bathe, MD, FRCSC, FACS,a and Chad G. Ball, MD, MSC, FRCSC, FACSa,* a b
Department of Surgery, Cumming School of Medicine, University of Calgary, Calgary, Canada Department of Medicine, Cumming School of Medicine, University of Calgary, Calgary, Canada
article info
abstract
Article history:
Background: The single best diagnostic and staging test for pancreatic cancer remains a
Received 2 January 2015
contrast-enhanced computed tomography scan. It is frequently the only imaging test
Received in revised form
required before surgical resection for solid pancreatic lesions. Unfortunately, many pa-
27 March 2015
tients undergo additional testing that often delays definitive care.
Accepted 8 April 2015
Materials and methods: A retrospective review of all patients with solid pancreatic lesions
Available online xxx
concerning for adenocarcinoma referred to a high volume Hepato-Pancreato-Biliary (HPB) service over 4 y (2008e2012) was completed. The time intervals between the initial imaging
Keywords:
test and both consultation with HPB surgery and operative intervention, as well as the
Pancreatic cancer
number of additional tests, were evaluated. Standard statistical methodology was used
Unnecessary testing
(P < 0.05). Results: Among 130 patients with solid pancreatic lesions, the index imaging modality was ultrasonography and computed tomography for 75 (58%) and 52 (40%), respectively. Patients underwent a mean of 1.3 diagnostic tests after the index study and before consultation with HPB surgery (range: 0e5). There was a significant increase in time to HPB consultation and operative intervention with an increasing number of interval imaging tests. The mean time to surgical consultation and operation if 0 interval diagnostic tests were performed was 15.9 and 45.4 d, respectively. If four interval tests were conducted, the mean was 69.4 and 122.6 d, respectively. Sixty-two patients (48%) were initially referred to a nonsurgical service. The mean time to surgical consultation and operation if an intervening referral occurred was 36.6 and 66.8 d, respectively. This compares to 19.8 and 48.1 d, respectively, in cases of direct referral to an HPB surgeon. The mean number of diagnostic tests performed before HPB consultation if a nonsurgical referral occurred was 2.1 (versus 0.7 if direct HPB surgeon referral). Conclusions: Despite a relatively simple algorithm for the investigation of solid pancreatic lesions, considerable heterogeneity remains in how these patients are evaluated before referral to HPB surgery. As the number of investigations increases after the index imaging test, there is increasing delay to both surgical consultation and definitive intervention. Education is required to expedite care and mitigate excess diagnostic tests. ª 2015 Elsevier Inc. All rights reserved.
* Corresponding author. Department of Surgery, Cumming School of Medicine, University of Calgary, Foothills Medical Centre, 1403 e 29 Street NW, Calgary, Alberta T2N 2T9, Canada. Tel.: þ1 403 944 3417; fax: þ1 403 944 8799. E-mail address: [email protected] (C.G. Ball). 0022-4804/$ e see front matter ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jss.2015.04.026
2
1.
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 5 ) 1 e5
Introduction
Pancreatic cancer remains the fourth leading cause of cancer death in the United States and the eighth worldwide [1,2]. Approximately 45,000 individuals are diagnosed with an exocrine pancreatic cancer annually in the United States alone [1]. Overall, pancreatic cancer has a very poor prognosis with a 5-y survival rate of less than 5% [2,3]. Surgical resection provides the only potentially curative treatment modality. After pancreaticoduodenectomy, the 5-y survival rate approaches 25%e30% in individuals with lymph node negative disease [4,5]. Distal pancreatectomy also remains the treatment of choice for cancers located in the body and tail of the gland. As a result, individuals with suspected solid pancreatic tumors require prompt referral to a pancreatic specialist for consideration of surgical management [6,7]. Preoperative tests should be limited to those with the ability to influence subsequent clinical decisions. The current gold standard diagnostic and staging test for solid pancreatic tumors is a contrast-enhanced computed tomography (CT) scan. CT has been demonstrated to be superior to both ultrasonography and magnetic resonance imaging with regard to accuracy for predicting unresectable lesions, as well as in overall sensitivity [8e10]. As a result, CT is often the only imaging test required before surgical exploration. Unfortunately, a recent study by Cooper et al. [7] noted an average of three unnecessary diagnostic tests were ordered for each patient with a clearly resectable pancreatic tumor. This resulted in considerable added costs, a significant increase in the wait time for definitive surgical management, and an element of risk inherent to the procedure itself (i.e., complications and radiation exposure). The primary aim of this study was therefore to clarify the efficiency of the current variable diagnostic pathways for patients with solid pancreatic neoplasms in the time interval before referral to a HepatoPancreato-Biliary (HPB) surgeon.
2.
Materials and methods
2.1.
Patient selection
Patients referred to the HPB service at the Foothills Medical Center in Calgary between January 1, 2008 and December 31, 2012 with a solid pancreatic lesion concerning for a pancreatic adenocarcinoma on diagnostic imaging were included. The Foothills Medical Center serves as the only pancreatic surgical center for a catchment of approximately 3 million people (southern part of Alberta, south eastern British Columbia, and south western Saskatchewan). There are no additional HPB surgeons within this catchment area. As a result, the referrals are a mix of urban and rural (80%:20%), which is also reflective of the population distribution. It should be noted that access to CT imaging (unlike magnetic resonance imaging) in urban and rural centers is comparably rapid. Nonoperative candidates, including those with metastatic disease, were also included. Data were retrospectively collected and reviewed using both electronic medical records and paper-based charts. Patients with a cystic mass or
diagnostic imaging not consistent with an adenocarcinoma were excluded.
2.2.
Data collection and statistical analysis
For each patient, the index diagnostic imaging test was identified. This was defined as the initial imaging test of any modality, from which concern for a pancreatic adenocarcinoma was reported by a faculty radiologist. All subsequent diagnostic procedures were numbered from this point of reference (time point zero). The specific time point of referral to the HPB service, as well as to medical oncology, the palliative care service, and/or operative intervention were also noted. Complications secondary to diagnostic procedures, as well as the number of referrals to nonsurgeon physicians after the index imaging test but before consultation by an HPB surgeon, was noted. Descriptive statistics were used, and t-test analyses were performed where appropriate (P < 0.05 ¼ significant). All P values are two tailed.
3.
Results
A total of 130 patients with imaging suggestive of a pancreatic adenocarcinoma were referred to our HPB service during the study interval. Fifty-eight percent of patients were female, and the mean age at the time of index imaging test was 63.3 y (range ¼ 36e86). The index imaging modality was ultrasonography for 75 (58%) and CT for 52 patients (40%), respectively. Three patients had pancreatic cancer identified incidentally on nonenhanced CT, CT-angiography, and CT-Kidney-Ureter-Bladder. Patients underwent a mean of 1.3 additional diagnostic tests after the index study and before consultation with an HPB surgeon (range ¼ 0e5 tests). Of the 75 individuals who underwent ultrasound as the initial imaging test, only 6 were referred to HPB surgery before undergoing a CT (5 went on to have a CT after their HPB consultation and 1 underwent an endoscopic ultrasound with biopsy and palliative care). Twenty-two percent of individuals were referred directly to HPB surgery after their index diagnostic test (i.e., 0 additional tests), whereas 42% and 24% underwent 1 or 2 additional tests, respectively. There was a significant increase in both the time to HPB consultation and operative intervention with increases in the number of interval imaging tests (Table 1, Figure; P < 0.05). The mean time to surgical consultation and operative intervention if 0 additional tests were performed was 15.9 and 45.4 d, respectively. In contrast, if 4 additional tests were conducted, the mean time was 69.4 d to HPB consultation and 122.6 d to an operation (P < 0.05). Comparisons in delay were also significant when comparing the 0 additional test group to the 2, 3, and 4 test groups (P < 0.05). A total of 62 patients were referred to a nonsurgeon specialist after the index imaging test but before consultation with HPB surgery. Fifty-four patients were met in consultation by gastroenterology, seven were reviewed by the hospitalist service, and five were sent to the general internal medicine service. Of note, 44 of the patients (81%) referred to gastroenterology underwent an endoscopic retrograde cholangiopancreatography (ERCP). The mean time from the index
3
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 5 ) 1 e5
Table 1 e Frequency distribution of delays after the index imaging test but before HPB consultation and date of operation. Number of tests 0 1 2 3 4 5
n (%) 28 55 31 7 8 1
(21.5) (42.3) (23.8) (5.4) (6.2) (0.8)
Days to HPB
Days to OR
N to OR
15.9 21.1 34.3 46.4 69.4 47
45.4 53.4 61.0 81.4 122.6 d
19 32 18 7 5 0
N ¼ number of patients; OR ¼ operating room.
imaging test to HPB consultation and operation if no interval consultation took place was 19.8 and 48.1 d, respectively. However, if an interval consultation to a nonsurgical specialist was triggered, these delays increased to 36.6 and 66.8 d, respectively (P < 0.05). Additionally, the mean number of diagnostic investigations performed before HPB consultation tripled from 0.7e2.1 if a consultation took place before HPB surgery evaluation (P < 0.05). As the number of tests increased, the proportion of patients undergoing surgery of any type was unchanged (P > 0.05). As the number of tests increased, the proportion of patients undergoing palliative (i.e., nonresective) surgery increased however (P < 0.05). A total of 74 patients (57%) underwent an ERCP procedure. As outlined, 44 patients had this test performed before HPB consultation, whereas 30 patients had an ERCP after their clinic visit. Only six patients experienced a complication secondary to ERCP (four cholangitis, one pancreatitis, and one retroperitoneal duodenal perforation). Ultimately, 79 of the patients (61%) referred with a suspected pancreatic adenocarcinoma underwent an operation. Most patients received a pancreaticoduodenectomy (Table 2); however, a significant number were managed with a palliative procedure (cancer deemed unresectable at operation). These palliative procedures included biliodigestive double bypass, gastrojejunostomy, biopsy with celiac plexus block, and diagnostic laparoscopy.
4.
Discussion
The clinical signs and symptoms of pancreatic cancer are clearly not specific enough to allow a definitive diagnosis based
Figure e Mean number of days to HPB surgery consultation and surgical intervention from index imaging test. Error bars represent the standard error.
Table 2 e Operation and pathology characteristics. Characteristic Total patients Surgery Pancreaticoduodenectomy Distal pancreatectomy/splenectomy Distal pancreatectomy/spleen preserving Biliodigestive double bypass Gastrojejunostomy Biopsy/celiac plexus block Diagnostic laparoscopy Surgical pathology Adenocarcinoma Neuroendocrine Chronic pancreatitis Autoimmune pancreatitis Solid pseudopapillary tumor Cholangiocarcinoma Metastatic RCC
n 130 79 34 9 6 16 5 6 2 63 8 4 1 1 1 1
RCC ¼ Renal Cell Cancer.
on history and/or physical examination alone. Patient presentations are variable and often vague, necessitating the use of diagnostic imaging procedures to confirm a suspected diagnosis. Transabdominal ultrasound remains the most common initial imaging modalities for suspected HPB pathology. Using ultrasound as a first-line imaging modality is both cost effective and supported in the literature where the sensitivity for detecting exocrine pancreatic cancer >3 cm has been demonstrated to be up to 90%, with a specificity of 99% [11e13]. Unfortunately, ultrasound performance suffers significantly for smaller tumors. The influence of body habitus, retroperitoneal positioning of the pancreas, and operator ability also prevent ultrasound from providing accurate staging information [14]. Nonetheless, the popularity of this initial diagnostic approach was confirmed in our study where 58% of patients underwent a transabdominal ultrasound that ultimately revealed concern for a solid pancreatic adenocarcinoma. Although ultrasound may be the most common first-line imaging, abdominal CT remains the most used modality for diagnosing and staging suspected pancreatic cancer. CT scanners have undergone significant advances over the past decade that include, but are not limited to, speed, ease of use, and reduced exposure to ionizing radiation. In a recent review of the literature, Shrikhande et al. [14] reveal that the sensitivity of CT for identifying pancreatic cancer is 75%e 100% with a specificity between 70% and 100%. CT also provides exceptional evaluation for resectability based on both locoregional and distant factors. The accuracy of CT in predicting vascular involvement is also excellent (81%e99%) [14e16]. As a result, CT remains the single best diagnostic and staging test for pancreatic cancer and is the only preoperative imaging test required before surgical intervention in resectable pancreatic cancer (i.e., surgical planning). Predictably in this study, all patients except one underwent CT imaging of the abdomen. Although ERCP is known to posses a high sensitivity and specificity for the diagnosis of pancreatic cancer [17], it is also associated with increased cost as well as risk to the patient given its invasive nature. A large multivariate analysis of over
4
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 5 ) 1 e5
11,000 patients who underwent ERCP revealed a complication rate approximating 4% (most commonly pancreatitis and hemorrhage) [18]. As a result, the role of ERCP in the pathway for suspected pancreatic cancer has transitioned from a primarily diagnostic to therapeutic endeavor. Unfortunately, because of prolonged wait, operative wait times, and significant patient discomfort associated with cholestasis, ERCP remains essential in many scenarios. A recent review demonstrated that preoperative biliary stenting doubled between 1992 and 2007, with most stenting occurring before surgical evaluation [19]. Numerous retrospective reviews and meta-analyses evaluating stenting (preoperative biliary instrumentation) in pancreatic cancer have revealed either no difference in outcome and/or increased complications [20e24]. Additionally, a randomized control trial in 2012 by Van Der Gaag et al. [25] reported that routine preoperative biliary drainage was associated with a significant increase in postoperative infectious complications. Predictably, preoperative stenting was also associated with a delay to surgical intervention [19]. The utility of any preoperative test is determined by its ability to directly influence clinical decision making. This principle should be carefully considered given the associated cost and exposure of the patient to potential risk when ordering any diagnostic test or procedure. In our study, patients underwent an average of 1.3 tests after the index imaging procedure, which identified concern for a solid pancreatic adenocarcinoma, but before HPB surgical consultation. Therefore, a mean of 2.3 tests were performed in total. As the number of tests increased, the time to surgical consultation and operative intervention also lengthened. The average time to surgical consultation and operation when no interval diagnostic tests were performed was 15.9 and 45.4 d, respectively. In contrast, if four interval tests were conducted, the average was 69.4 and 122.6 d, respectively. These scenarios have been confirmed in the literature to result in up to three-fold increases in delays to operative intervention [19]. This pattern also provides us with a strong reminder that given the general lack of therapeutic options with a potential for cure, these patients should be promptly referred to an HPB surgical service for consideration of operative management. Furthermore, although the delay to surgical intervention increased with additional consultations in the pre-HPB surgeon referral interval, it is unclear if this had any truly relevant tumor-related impact beyond the likely patient-centered psychological strain associated with wait times. Unfortunately, the delays associated with ERCP are much more complicated and continue to surround referral patterns and even more importantly operative capacity and throughput (i.e., delayed access to the operating room). When taken as a whole, these data do not “definitely” confirm anecdotal concerns that delays arising from either noncontributory diagnostic testing and/or consultation with nonsurgeon specialists adversely influence patient outcomes. The observation that significantly more patients received palliative operative interventions (rather than potentially curative resections) as the number of interval diagnostic tests increased is highly concerning and requires further investigation however.
5.
Conclusions
In summary, despite a simple algorithm for the workup of solid pancreatic lesions that generally requires only a preoperative CT, considerable heterogeneity remains in patient pathways before referral to HPB surgery. As the number of investigations or consultations increases after the index imaging test, there are increasing delays to both surgical consultation and definitive operative intervention, as well as an increase in the proportion of patients receiving palliative procedures. Further education is required to expedite care and mitigate excess diagnostic tests. This includes focusing on earlier referral to a high-volume HPB surgeon to assist in guiding investigations, as well as making patient-centered decisions based on rapid CT imaging.
Acknowledgment This study was supported in part by a grant from the Calgary Surgery Research Development Fund. Authors’ contributions: M.R.D. contributed to the ethics proposal, study design, data collection, data analysis, and is the primary author of the abstract and article. C.G.B. is a senior author, who provided assistance with the ethics proposal, study design, data analysis, and writing of the abstract and article. E.D., R.M., F.R.S., and O.F.B. provided assistance with authorship of the article.
Disclosure The authors report no conflict of interest associated with any of the topics discussed in this article.
references
[1] Siegel R, Ma J, Zou Z, Jemal A. Cancer statistics, 2014. CA Cancer J Clin 2014;64:9. [2] Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA Cancer J Clin 2011;61:69. [3] Li D, Xie K, Wolff R, Abbruzzese JL. Pancreatic cancer. Lancet 2004;363:1049. [4] Trede M, Schwall G, Saeger HD. Survival after pancreatoduodenectomy. 118 consecutive resections without an operative mortality. Ann Surg 1990;211:447. [5] Cameron JL, Riall TS, Coleman J, Belcher KA. One thousand consecutive pancreaticoduodenectomies. Ann Surg 2006; 244:10. [6] Vincent A, Herman J, Schulick R, Hruban RH, Goggins M. Pancreatic cancer. Lancet 2011;378:607. [7] Cooper M, Newman NA, Ibrahim AM, et al. Unnecessary tests and procedures in patients presenting with solid tumors of the pancreas. J Gastrointest Surg 2013;17:1218. [8] Bipat S, Phoa SS, van Delden OM, et al. Ultrasonography, computed tomography and magnetic resonance imaging for diagnosis and determining resectability of pancreatic adenocarcinoma: a meta-analysis. J Comput Assist Tomogr 2005;29:438.
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 5 ) 1 e5
[9] Squillaci E, Fanucci E, Sciuto F, et al. Vascular involvement in pancreatic neoplasm: a comparison between spiral CT and DSA. Dig Dis Sci 2003;48:449. [10] Valls C, Andia E, Sanchez A, et al. Dual-phase helical CT of pancreatic adenocarcinoma: assessment of resectability before surgery. AJR Am J Roentgenol 2002;178:821. [11] Karlson BM, Ekbom A, Lindgren PG, Kallskog V, Rastad J. Abdominal US for diagnosis of pancreatic tumor: prospective cohort analysis. Radiology 1999;213:107. [12] Brambs HJ, Claussen CD. Pancreatic and ampullary carcinoma. ultrasound, computed tomography, magnetic resonance imaging and angiography. Endoscopy 1993;25:58. [13] Maringhini A, Ciambra M, Raimondo M, et al. Clinical presentation and ultrasonography in the diagnosis of pancreatic cancer. Pancreas 1993;8:146. [14] Shrikhande SV, Barreto SG, Goel M, Arya S. Multimodality imaging of pancreatic ductal adenocarcinoma: a review of the literature. HPB 2012;14:658. [15] Imbriaco M, Megibow AJ, Ragozzino A, et al. Value of the single-phase technique in MDCT assessment of pancreatic tumors. AJR Am J Roentgenol 2005;184:1111. [16] Vargas R, Nino-Murcia M, Trueblood W, Jeffrey RB Jr. MDCT in pancreatic adenocarcinoma: prediction of vascular invasion and resectability using a multiphasic technique with curved planar reformations. AJR Am J Roentgenol 2004; 182:419. [17] Niederau C, Grendell JH. Diagnosis of pancreatic carcinoma. imaging techniques and tumor markers. Pancreas 1992;7:66.
5
[18] Cotton PB, Garrow DA, Gallagher J, Romagnuolo J. Risk factors for complications after ERCP: a multivariate analysis of 11,497 procedures over 12 years. Gastrointest Endosc 2009; 70:80. [19] Jinkins LJ, Parmar AD, Han Y, et al. Current trends in preoperative biliary stenting in patients with pancreatic cancer. Surgery 2013;154:179. [20] Tsai YF, Shyu JF, Chen TH, Shyr YM, Su CH. Effect of preoperative biliary drainage on surgical outcome after pancreaticoduodenectomy. Hepatogastroenterology 2006;53: 823. [21] Jagannath P, Dhir V, Shrikhande S, Shah RC, Mullerpatan P, Mohandas KM. Effect of preoperative biliary stenting on immediate outcome after pancreaticoduodenectomy. The Br J Surg 2005;92:356. [22] Sewnath ME, Birjmohun RS, Rauws EA, Huibregtse K, Obertop H, Gouma DJ. The effect of preoperative biliary drainage on postoperative complications after pancreaticoduodenectomy. J Am Coll Surg 2001;192:726. [23] Garcea G, Chee W, Ong SL, Maddern GJ. Preoperative biliary drainage for distal obstruction: the case against revisited. Pancreas 2010;39:119. [24] Qiu YD, Bai JL, Xu FG, Ding YT. Effect of preoperative biliary drainage on malignant obstructive jaundice: a metaanalysis. World J Gastroenterol 2011;17:391. [25] Van der Gaag NA, Rauws EA, van Eijck CH, et al. Preoperative biliary drainage for cancer of the head of the pancreas. The New Engl J Med 2010;362:129.
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.JournalofSurgicalResearch.com
The diagnostic pathway for solid pancreatic neoplasms: are we applying too many tests? Michael R. Driedger, MD,a Elijah Dixon, MD, MSC, FRCSC, FACS,a Rachid Mohamed, MD, FRCPC,b Francis R. Sutherland, MD, FRCSC,a Oliver F. Bathe, MD, FRCSC, FACS,a and Chad G. Ball, MD, MSC, FRCSC, FACSa,* a b
Department of Surgery, Cumming School of Medicine, University of Calgary, Calgary, Canada Department of Medicine, Cumming School of Medicine, University of Calgary, Calgary, Canada
article info
abstract
Article history:
Background: The single best diagnostic and staging test for pancreatic cancer remains a
Received 2 January 2015
contrast-enhanced computed tomography scan. It is frequently the only imaging test
Received in revised form
required before surgical resection for solid pancreatic lesions. Unfortunately, many pa-
27 March 2015
tients undergo additional testing that often delays definitive care.
Accepted 8 April 2015
Materials and methods: A retrospective review of all patients with solid pancreatic lesions
Available online xxx
concerning for adenocarcinoma referred to a high volume Hepato-Pancreato-Biliary (HPB) service over 4 y (2008e2012) was completed. The time intervals between the initial imaging
Keywords:
test and both consultation with HPB surgery and operative intervention, as well as the
Pancreatic cancer
number of additional tests, were evaluated. Standard statistical methodology was used
Unnecessary testing
(P < 0.05). Results: Among 130 patients with solid pancreatic lesions, the index imaging modality was ultrasonography and computed tomography for 75 (58%) and 52 (40%), respectively. Patients underwent a mean of 1.3 diagnostic tests after the index study and before consultation with HPB surgery (range: 0e5). There was a significant increase in time to HPB consultation and operative intervention with an increasing number of interval imaging tests. The mean time to surgical consultation and operation if 0 interval diagnostic tests were performed was 15.9 and 45.4 d, respectively. If four interval tests were conducted, the mean was 69.4 and 122.6 d, respectively. Sixty-two patients (48%) were initially referred to a nonsurgical service. The mean time to surgical consultation and operation if an intervening referral occurred was 36.6 and 66.8 d, respectively. This compares to 19.8 and 48.1 d, respectively, in cases of direct referral to an HPB surgeon. The mean number of diagnostic tests performed before HPB consultation if a nonsurgical referral occurred was 2.1 (versus 0.7 if direct HPB surgeon referral). Conclusions: Despite a relatively simple algorithm for the investigation of solid pancreatic lesions, considerable heterogeneity remains in how these patients are evaluated before referral to HPB surgery. As the number of investigations increases after the index imaging test, there is increasing delay to both surgical consultation and definitive intervention. Education is required to expedite care and mitigate excess diagnostic tests. ª 2015 Elsevier Inc. All rights reserved.
* Corresponding author. Department of Surgery, Cumming School of Medicine, University of Calgary, Foothills Medical Centre, 1403 e 29 Street NW, Calgary, Alberta T2N 2T9, Canada. Tel.: þ1 403 944 3417; fax: þ1 403 944 8799. E-mail address: [email protected] (C.G. Ball). 0022-4804/$ e see front matter ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jss.2015.04.026
2
1.
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 5 ) 1 e5
Introduction
Pancreatic cancer remains the fourth leading cause of cancer death in the United States and the eighth worldwide [1,2]. Approximately 45,000 individuals are diagnosed with an exocrine pancreatic cancer annually in the United States alone [1]. Overall, pancreatic cancer has a very poor prognosis with a 5-y survival rate of less than 5% [2,3]. Surgical resection provides the only potentially curative treatment modality. After pancreaticoduodenectomy, the 5-y survival rate approaches 25%e30% in individuals with lymph node negative disease [4,5]. Distal pancreatectomy also remains the treatment of choice for cancers located in the body and tail of the gland. As a result, individuals with suspected solid pancreatic tumors require prompt referral to a pancreatic specialist for consideration of surgical management [6,7]. Preoperative tests should be limited to those with the ability to influence subsequent clinical decisions. The current gold standard diagnostic and staging test for solid pancreatic tumors is a contrast-enhanced computed tomography (CT) scan. CT has been demonstrated to be superior to both ultrasonography and magnetic resonance imaging with regard to accuracy for predicting unresectable lesions, as well as in overall sensitivity [8e10]. As a result, CT is often the only imaging test required before surgical exploration. Unfortunately, a recent study by Cooper et al. [7] noted an average of three unnecessary diagnostic tests were ordered for each patient with a clearly resectable pancreatic tumor. This resulted in considerable added costs, a significant increase in the wait time for definitive surgical management, and an element of risk inherent to the procedure itself (i.e., complications and radiation exposure). The primary aim of this study was therefore to clarify the efficiency of the current variable diagnostic pathways for patients with solid pancreatic neoplasms in the time interval before referral to a HepatoPancreato-Biliary (HPB) surgeon.
2.
Materials and methods
2.1.
Patient selection
Patients referred to the HPB service at the Foothills Medical Center in Calgary between January 1, 2008 and December 31, 2012 with a solid pancreatic lesion concerning for a pancreatic adenocarcinoma on diagnostic imaging were included. The Foothills Medical Center serves as the only pancreatic surgical center for a catchment of approximately 3 million people (southern part of Alberta, south eastern British Columbia, and south western Saskatchewan). There are no additional HPB surgeons within this catchment area. As a result, the referrals are a mix of urban and rural (80%:20%), which is also reflective of the population distribution. It should be noted that access to CT imaging (unlike magnetic resonance imaging) in urban and rural centers is comparably rapid. Nonoperative candidates, including those with metastatic disease, were also included. Data were retrospectively collected and reviewed using both electronic medical records and paper-based charts. Patients with a cystic mass or
diagnostic imaging not consistent with an adenocarcinoma were excluded.
2.2.
Data collection and statistical analysis
For each patient, the index diagnostic imaging test was identified. This was defined as the initial imaging test of any modality, from which concern for a pancreatic adenocarcinoma was reported by a faculty radiologist. All subsequent diagnostic procedures were numbered from this point of reference (time point zero). The specific time point of referral to the HPB service, as well as to medical oncology, the palliative care service, and/or operative intervention were also noted. Complications secondary to diagnostic procedures, as well as the number of referrals to nonsurgeon physicians after the index imaging test but before consultation by an HPB surgeon, was noted. Descriptive statistics were used, and t-test analyses were performed where appropriate (P < 0.05 ¼ significant). All P values are two tailed.
3.
Results
A total of 130 patients with imaging suggestive of a pancreatic adenocarcinoma were referred to our HPB service during the study interval. Fifty-eight percent of patients were female, and the mean age at the time of index imaging test was 63.3 y (range ¼ 36e86). The index imaging modality was ultrasonography for 75 (58%) and CT for 52 patients (40%), respectively. Three patients had pancreatic cancer identified incidentally on nonenhanced CT, CT-angiography, and CT-Kidney-Ureter-Bladder. Patients underwent a mean of 1.3 additional diagnostic tests after the index study and before consultation with an HPB surgeon (range ¼ 0e5 tests). Of the 75 individuals who underwent ultrasound as the initial imaging test, only 6 were referred to HPB surgery before undergoing a CT (5 went on to have a CT after their HPB consultation and 1 underwent an endoscopic ultrasound with biopsy and palliative care). Twenty-two percent of individuals were referred directly to HPB surgery after their index diagnostic test (i.e., 0 additional tests), whereas 42% and 24% underwent 1 or 2 additional tests, respectively. There was a significant increase in both the time to HPB consultation and operative intervention with increases in the number of interval imaging tests (Table 1, Figure; P < 0.05). The mean time to surgical consultation and operative intervention if 0 additional tests were performed was 15.9 and 45.4 d, respectively. In contrast, if 4 additional tests were conducted, the mean time was 69.4 d to HPB consultation and 122.6 d to an operation (P < 0.05). Comparisons in delay were also significant when comparing the 0 additional test group to the 2, 3, and 4 test groups (P < 0.05). A total of 62 patients were referred to a nonsurgeon specialist after the index imaging test but before consultation with HPB surgery. Fifty-four patients were met in consultation by gastroenterology, seven were reviewed by the hospitalist service, and five were sent to the general internal medicine service. Of note, 44 of the patients (81%) referred to gastroenterology underwent an endoscopic retrograde cholangiopancreatography (ERCP). The mean time from the index
3
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 5 ) 1 e5
Table 1 e Frequency distribution of delays after the index imaging test but before HPB consultation and date of operation. Number of tests 0 1 2 3 4 5
n (%) 28 55 31 7 8 1
(21.5) (42.3) (23.8) (5.4) (6.2) (0.8)
Days to HPB
Days to OR
N to OR
15.9 21.1 34.3 46.4 69.4 47
45.4 53.4 61.0 81.4 122.6 d
19 32 18 7 5 0
N ¼ number of patients; OR ¼ operating room.
imaging test to HPB consultation and operation if no interval consultation took place was 19.8 and 48.1 d, respectively. However, if an interval consultation to a nonsurgical specialist was triggered, these delays increased to 36.6 and 66.8 d, respectively (P < 0.05). Additionally, the mean number of diagnostic investigations performed before HPB consultation tripled from 0.7e2.1 if a consultation took place before HPB surgery evaluation (P < 0.05). As the number of tests increased, the proportion of patients undergoing surgery of any type was unchanged (P > 0.05). As the number of tests increased, the proportion of patients undergoing palliative (i.e., nonresective) surgery increased however (P < 0.05). A total of 74 patients (57%) underwent an ERCP procedure. As outlined, 44 patients had this test performed before HPB consultation, whereas 30 patients had an ERCP after their clinic visit. Only six patients experienced a complication secondary to ERCP (four cholangitis, one pancreatitis, and one retroperitoneal duodenal perforation). Ultimately, 79 of the patients (61%) referred with a suspected pancreatic adenocarcinoma underwent an operation. Most patients received a pancreaticoduodenectomy (Table 2); however, a significant number were managed with a palliative procedure (cancer deemed unresectable at operation). These palliative procedures included biliodigestive double bypass, gastrojejunostomy, biopsy with celiac plexus block, and diagnostic laparoscopy.
4.
Discussion
The clinical signs and symptoms of pancreatic cancer are clearly not specific enough to allow a definitive diagnosis based
Figure e Mean number of days to HPB surgery consultation and surgical intervention from index imaging test. Error bars represent the standard error.
Table 2 e Operation and pathology characteristics. Characteristic Total patients Surgery Pancreaticoduodenectomy Distal pancreatectomy/splenectomy Distal pancreatectomy/spleen preserving Biliodigestive double bypass Gastrojejunostomy Biopsy/celiac plexus block Diagnostic laparoscopy Surgical pathology Adenocarcinoma Neuroendocrine Chronic pancreatitis Autoimmune pancreatitis Solid pseudopapillary tumor Cholangiocarcinoma Metastatic RCC
n 130 79 34 9 6 16 5 6 2 63 8 4 1 1 1 1
RCC ¼ Renal Cell Cancer.
on history and/or physical examination alone. Patient presentations are variable and often vague, necessitating the use of diagnostic imaging procedures to confirm a suspected diagnosis. Transabdominal ultrasound remains the most common initial imaging modalities for suspected HPB pathology. Using ultrasound as a first-line imaging modality is both cost effective and supported in the literature where the sensitivity for detecting exocrine pancreatic cancer >3 cm has been demonstrated to be up to 90%, with a specificity of 99% [11e13]. Unfortunately, ultrasound performance suffers significantly for smaller tumors. The influence of body habitus, retroperitoneal positioning of the pancreas, and operator ability also prevent ultrasound from providing accurate staging information [14]. Nonetheless, the popularity of this initial diagnostic approach was confirmed in our study where 58% of patients underwent a transabdominal ultrasound that ultimately revealed concern for a solid pancreatic adenocarcinoma. Although ultrasound may be the most common first-line imaging, abdominal CT remains the most used modality for diagnosing and staging suspected pancreatic cancer. CT scanners have undergone significant advances over the past decade that include, but are not limited to, speed, ease of use, and reduced exposure to ionizing radiation. In a recent review of the literature, Shrikhande et al. [14] reveal that the sensitivity of CT for identifying pancreatic cancer is 75%e 100% with a specificity between 70% and 100%. CT also provides exceptional evaluation for resectability based on both locoregional and distant factors. The accuracy of CT in predicting vascular involvement is also excellent (81%e99%) [14e16]. As a result, CT remains the single best diagnostic and staging test for pancreatic cancer and is the only preoperative imaging test required before surgical intervention in resectable pancreatic cancer (i.e., surgical planning). Predictably in this study, all patients except one underwent CT imaging of the abdomen. Although ERCP is known to posses a high sensitivity and specificity for the diagnosis of pancreatic cancer [17], it is also associated with increased cost as well as risk to the patient given its invasive nature. A large multivariate analysis of over
4
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 5 ) 1 e5
11,000 patients who underwent ERCP revealed a complication rate approximating 4% (most commonly pancreatitis and hemorrhage) [18]. As a result, the role of ERCP in the pathway for suspected pancreatic cancer has transitioned from a primarily diagnostic to therapeutic endeavor. Unfortunately, because of prolonged wait, operative wait times, and significant patient discomfort associated with cholestasis, ERCP remains essential in many scenarios. A recent review demonstrated that preoperative biliary stenting doubled between 1992 and 2007, with most stenting occurring before surgical evaluation [19]. Numerous retrospective reviews and meta-analyses evaluating stenting (preoperative biliary instrumentation) in pancreatic cancer have revealed either no difference in outcome and/or increased complications [20e24]. Additionally, a randomized control trial in 2012 by Van Der Gaag et al. [25] reported that routine preoperative biliary drainage was associated with a significant increase in postoperative infectious complications. Predictably, preoperative stenting was also associated with a delay to surgical intervention [19]. The utility of any preoperative test is determined by its ability to directly influence clinical decision making. This principle should be carefully considered given the associated cost and exposure of the patient to potential risk when ordering any diagnostic test or procedure. In our study, patients underwent an average of 1.3 tests after the index imaging procedure, which identified concern for a solid pancreatic adenocarcinoma, but before HPB surgical consultation. Therefore, a mean of 2.3 tests were performed in total. As the number of tests increased, the time to surgical consultation and operative intervention also lengthened. The average time to surgical consultation and operation when no interval diagnostic tests were performed was 15.9 and 45.4 d, respectively. In contrast, if four interval tests were conducted, the average was 69.4 and 122.6 d, respectively. These scenarios have been confirmed in the literature to result in up to three-fold increases in delays to operative intervention [19]. This pattern also provides us with a strong reminder that given the general lack of therapeutic options with a potential for cure, these patients should be promptly referred to an HPB surgical service for consideration of operative management. Furthermore, although the delay to surgical intervention increased with additional consultations in the pre-HPB surgeon referral interval, it is unclear if this had any truly relevant tumor-related impact beyond the likely patient-centered psychological strain associated with wait times. Unfortunately, the delays associated with ERCP are much more complicated and continue to surround referral patterns and even more importantly operative capacity and throughput (i.e., delayed access to the operating room). When taken as a whole, these data do not “definitely” confirm anecdotal concerns that delays arising from either noncontributory diagnostic testing and/or consultation with nonsurgeon specialists adversely influence patient outcomes. The observation that significantly more patients received palliative operative interventions (rather than potentially curative resections) as the number of interval diagnostic tests increased is highly concerning and requires further investigation however.
5.
Conclusions
In summary, despite a simple algorithm for the workup of solid pancreatic lesions that generally requires only a preoperative CT, considerable heterogeneity remains in patient pathways before referral to HPB surgery. As the number of investigations or consultations increases after the index imaging test, there are increasing delays to both surgical consultation and definitive operative intervention, as well as an increase in the proportion of patients receiving palliative procedures. Further education is required to expedite care and mitigate excess diagnostic tests. This includes focusing on earlier referral to a high-volume HPB surgeon to assist in guiding investigations, as well as making patient-centered decisions based on rapid CT imaging.
Acknowledgment This study was supported in part by a grant from the Calgary Surgery Research Development Fund. Authors’ contributions: M.R.D. contributed to the ethics proposal, study design, data collection, data analysis, and is the primary author of the abstract and article. C.G.B. is a senior author, who provided assistance with the ethics proposal, study design, data analysis, and writing of the abstract and article. E.D., R.M., F.R.S., and O.F.B. provided assistance with authorship of the article.
Disclosure The authors report no conflict of interest associated with any of the topics discussed in this article.
references
[1] Siegel R, Ma J, Zou Z, Jemal A. Cancer statistics, 2014. CA Cancer J Clin 2014;64:9. [2] Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA Cancer J Clin 2011;61:69. [3] Li D, Xie K, Wolff R, Abbruzzese JL. Pancreatic cancer. Lancet 2004;363:1049. [4] Trede M, Schwall G, Saeger HD. Survival after pancreatoduodenectomy. 118 consecutive resections without an operative mortality. Ann Surg 1990;211:447. [5] Cameron JL, Riall TS, Coleman J, Belcher KA. One thousand consecutive pancreaticoduodenectomies. Ann Surg 2006; 244:10. [6] Vincent A, Herman J, Schulick R, Hruban RH, Goggins M. Pancreatic cancer. Lancet 2011;378:607. [7] Cooper M, Newman NA, Ibrahim AM, et al. Unnecessary tests and procedures in patients presenting with solid tumors of the pancreas. J Gastrointest Surg 2013;17:1218. [8] Bipat S, Phoa SS, van Delden OM, et al. Ultrasonography, computed tomography and magnetic resonance imaging for diagnosis and determining resectability of pancreatic adenocarcinoma: a meta-analysis. J Comput Assist Tomogr 2005;29:438.
j o u r n a l o f s u r g i c a l r e s e a r c h x x x ( 2 0 1 5 ) 1 e5
[9] Squillaci E, Fanucci E, Sciuto F, et al. Vascular involvement in pancreatic neoplasm: a comparison between spiral CT and DSA. Dig Dis Sci 2003;48:449. [10] Valls C, Andia E, Sanchez A, et al. Dual-phase helical CT of pancreatic adenocarcinoma: assessment of resectability before surgery. AJR Am J Roentgenol 2002;178:821. [11] Karlson BM, Ekbom A, Lindgren PG, Kallskog V, Rastad J. Abdominal US for diagnosis of pancreatic tumor: prospective cohort analysis. Radiology 1999;213:107. [12] Brambs HJ, Claussen CD. Pancreatic and ampullary carcinoma. ultrasound, computed tomography, magnetic resonance imaging and angiography. Endoscopy 1993;25:58. [13] Maringhini A, Ciambra M, Raimondo M, et al. Clinical presentation and ultrasonography in the diagnosis of pancreatic cancer. Pancreas 1993;8:146. [14] Shrikhande SV, Barreto SG, Goel M, Arya S. Multimodality imaging of pancreatic ductal adenocarcinoma: a review of the literature. HPB 2012;14:658. [15] Imbriaco M, Megibow AJ, Ragozzino A, et al. Value of the single-phase technique in MDCT assessment of pancreatic tumors. AJR Am J Roentgenol 2005;184:1111. [16] Vargas R, Nino-Murcia M, Trueblood W, Jeffrey RB Jr. MDCT in pancreatic adenocarcinoma: prediction of vascular invasion and resectability using a multiphasic technique with curved planar reformations. AJR Am J Roentgenol 2004; 182:419. [17] Niederau C, Grendell JH. Diagnosis of pancreatic carcinoma. imaging techniques and tumor markers. Pancreas 1992;7:66.
5
[18] Cotton PB, Garrow DA, Gallagher J, Romagnuolo J. Risk factors for complications after ERCP: a multivariate analysis of 11,497 procedures over 12 years. Gastrointest Endosc 2009; 70:80. [19] Jinkins LJ, Parmar AD, Han Y, et al. Current trends in preoperative biliary stenting in patients with pancreatic cancer. Surgery 2013;154:179. [20] Tsai YF, Shyu JF, Chen TH, Shyr YM, Su CH. Effect of preoperative biliary drainage on surgical outcome after pancreaticoduodenectomy. Hepatogastroenterology 2006;53: 823. [21] Jagannath P, Dhir V, Shrikhande S, Shah RC, Mullerpatan P, Mohandas KM. Effect of preoperative biliary stenting on immediate outcome after pancreaticoduodenectomy. The Br J Surg 2005;92:356. [22] Sewnath ME, Birjmohun RS, Rauws EA, Huibregtse K, Obertop H, Gouma DJ. The effect of preoperative biliary drainage on postoperative complications after pancreaticoduodenectomy. J Am Coll Surg 2001;192:726. [23] Garcea G, Chee W, Ong SL, Maddern GJ. Preoperative biliary drainage for distal obstruction: the case against revisited. Pancreas 2010;39:119. [24] Qiu YD, Bai JL, Xu FG, Ding YT. Effect of preoperative biliary drainage on malignant obstructive jaundice: a metaanalysis. World J Gastroenterol 2011;17:391. [25] Van der Gaag NA, Rauws EA, van Eijck CH, et al. Preoperative biliary drainage for cancer of the head of the pancreas. The New Engl J Med 2010;362:129.
