Abstract
Background & Aims
Chronic pancreatitis is characterized by inflammation, atrophy, fibrosis with progressive ductal changes, and functional changes that include variable exocrine and endocrine insufficiency and multiple patterns of pain. We investigated whether abdominal imaging features accurately predict patterns of pain.
Methods
We collected data from participants in North American Pancreatitis Study 2 Continuation and Validation, a prospective multicenter study of patients with chronic pancreatitis performed at 13 expert centers in the United States from July 2008 through March 2012. Chronic pancreatitis was defined based on detection of characteristic changes by cross-sectional abdominal imaging, endoscopic retrograde cholangiopancreatography, endoscopic ultrasonography, or histology analyses. Patients were asked by a physician or trained clinical research coordinator if they had any abdominal pain in the year before enrollment; those that responded “yes” were asked to select from a list of 5 pain patterns. Using these patterns, we classified patients’ pain based on timing and severity. Abnormal pancreatitis-associated features on abdominal imaging were recorded using standardized case report forms.
Results
Data were collected from 518 patients (mean age, 52±14.6 years; 55% male; and 87.6% white). The most common physician-identified etiologies were alcohol (45.8%) and idiopathic (24.3%); 15.6% of patients reported no abdominal pain in the year before enrollment. The most common individual pain pattern was described as constant mild pain with episodes of severe pain, reported in 45% of patients. The most common imaging findings included pancreatic ductal dilatation (68%), atrophy (57%), and calcifications (55%). Imaging findings were categorized as obstructive for 20% and inflammatory for 25% of cases. The distribution of individual imaging findings was similar among patients with different patterns of pain. The distribution of pain patterns did not differ among clinically relevant groups of imaging findings.
Conclusions
Mechanisms that determine patterns and severity of pain in patients with chronic pancreatitis are largely independent of structural variants observed by abdominal imaging techniques. Pancreas-relevant quantitative and qualitative pain measures should be included in the evaluation of patients with chronic pancreatitis, to assess pain severity independently of imaging findings.
Keywords: NAPS 2-CV, chronic pancreatitis, abdominal pain, abdominal imaging
Abdominal pain remains one of the most distressing symptoms of chronic pancreatitis. The character of abdominal pain in these patients is highly variable with some having severe daily pain while others have fleeting discomfort or are asymptomatic (1). The pathogenesis of abdominal pain in chronic pancreatitis is poorly understood, likely due to multiple mechanisms, and variable mechanisms may be operative at different time points (2). Imaging studies are important to confirm the diagnosis as well as to assess for disease severity and complications. Cross sectional abdominal imaging remains the mainstay for evaluation of abdominal pain in chronic pancreatitis in anticipation of identifying structural abnormalities that can be treated with endoscopic or surgical approaches. In most patients, radiographic findings coupled with the patient's complaints help dictate subsequent management. A number of prior studies utilizing cross-sectional abdominal imaging, endoscopic retrograde cholangiopancreatography (ERCP), and endoscopic ultrasonography (EUS) have correlated ductal changes and morphologic abnormalities suggesting a poor correlation between these changes and abdominal pain (3-9). However, these studies typically evaluated pain severity, rather than pain pattern, which correlates poorly with pancreatitis-associated quality of life and disability (1).
Using data from a prospective study evaluating patients with chronic pancreatitis, which included details on pain patterns and imaging findings, we were able to assess the relationship between such radiographic abnormalities and the patient's complaints.
PATIENTS & METHODS
North American Pancreatitis Study 2 - Continuation and Validation Study (NAPS2-CV)
The NAPS2-CV study is a multicenter study that prospectively enrolled patients with chronic pancreatitis from 13 US centers with specific interest in pancreatic disease from July 2008 - March 2012. The NAPS2-CV study is part of the NAPS2 Program which has an overall goal to better understand the role of environmental and genetic factors in the susceptibility and progression of pancreatitis, and was developed to ascertain a genetic replication cohort for the NAPS2 cohort (10). The study was approved by the Institutional Review Board of individual participating centers and all study subjects signed an informed consent before enrollment.
The current paper focuses only on patients with chronic pancreatitis (n=521) enrolled in the NAPS2-CV study. Chronic pancreatitis was defined by the presence of characteristic changes on abdominal imaging studies (computerized tomography [CT] scan, magnetic resonance imaging [MRI]/Magnetic resonance cholangiopancreatography [MRCP], endoscopic retrograde cholangiopancreatography [ERCP] [Cambridge classification] or endoscopic ultrasound [EUS] [presence of 5 or more findings or presence of calcifications]) or histology. The number of patients fulfilling the entry criteria on one or more individual studies was – CT scan (n=323), ERCP (n=176), MRI/MRCP (n=194), EUS (n=132) and histology (n =25).
Patient and Physician Questionnaires
Information was collected from each patient and the enrolling physician using two sets of questionnaires. Patient questionnaires were administered by a physician or trained clinical research coordinator to collect detailed information on demographics, personal and family history, alcohol consumption, smoking, symptoms (including pain – see below), hospitalizations and emergency room visits, medication use and quality of life (SF-12 v2). Physicians answered questions relating to disease phenotype (history of AP, age of first acute pancreatitis attack, history of recurrent acute pancreatitis and number of acute pancreatitis attacks patients, age at onset of symptoms and diagnosis of chronic pancreatitis, presence and pattern of pain [see below], exocrine and endocrine insufficiency (presence, age of diagnosis, method of diagnosis, treatment), imaging findings [see below], histology, TIGAR-O risk factors (11), etiology, treatments tried and their perceived effectiveness.
Presence of pain and pain patterns
Patients were asked if they had any abdominal pain from chronic pancreatitis in the year before enrollment. Patients who responded “yes” to this question were further asked to choose the pattern of pain that best describes their pain experience from a list of five pain patterns (Table 1) (1).
Table 1.
Definitions and distribution of pain patterns, temporal nature and severity of pain and classification of imaging findings into hierarchical categories in 518 patients with chronic pancreatitis in the NAPS2-CV study
| Pain Pattern^ | Description |
|---|---|
| - (n=81, 15.6%) | No pain |
| A (n=67, 12.9%) | Usually pain free, but episodes of mild to moderate pain |
| B (n=23, 4.6%) | Constant mild to moderate pain |
| C (n=98, 18.9%) | Usually free of abdominal pain, but episodes of severe pain |
| D (n=229, 44.2%) | Constant mild to moderate pain plus episodes of severe pain |
| E (n-19, 3.7%) | Constant severe pain |
| Temporal Nature of pain^ | |
| Intermittent (n=165, 31.9%) | Pain pattern A or C |
| Constant (n=272, 52.5%) | Pain pattern B or D or E |
| Pain severity^ | |
| Mild-moderate (n=91, 17.6%) | Pain pattern A or B |
| Severe (n=346, 66.8%) | Pain pattern C or D or E |
| Imaging categories* | |
| Obstructive (n=103, 19.9%) | Presence of pancreatic duct obstruction/stricture irrespective of the presence of any additional imaging finding |
| Inflammation (n=130, 25.1%) | Presence of inflammation or inflammatory mass irrespective of the presence of any additional imaging finding |
| Pseudocyst (n=52, 10.0%) | Presence of pseudocyst(s) irrespective of the presence of any additional imaging finding |
| All others (n=233, 45.0%) | All remaining patients |
Based on patients description of pain experience in the year before study enrollment
Patients were assigned to an imaging category using a hierarchical model based on physician responses to the presence of individual imaging findings on the most recent studies. The categories are mutually exclusive. The hierarchical model starts with the obstructive category and assigns each subsequent category only to the remaining patients. A patient in a higher category can have findings from a subsequent category but not vice versa.
Patient responses were used to determine the temporal nature (intermittent vs. constant) and severity (mild-moderate vs. severe) of their pain experience. Patients were asked to give the frequency and duration of their pain attacks, to rate their pain experience (McGill pain questionnaire [12,13], PROMIS pain impact), and to specify their use of pain medications. Physicians were also asked to provide their interpretation of the patient's pain symptoms using the five categories of pain patterns, pain frequency and duration, and pain medication use. Since the focus of this manuscript is the correlation of pain experience from the patient's perspective with their findings on imaging studies, the case report form containing the physician's interpretation of the patient's pain experience was not used for analyses.
Imaging studies and Classification of findings
The enrolling physician was asked to indicate the study that established the diagnosis (imaging study, histology) and the finding used to diagnose chronic pancreatitis (e.g. calcifications, atrophy, etc.). Physicians provided information regarding the date and findings of the most recent imaging studies. In the initial half of the patient enrollment period, findings on individual studies were recorded; in the later half, presence of individual findings on any of the most recent imaging was recorded.
Data was collected for the following imaging findings – calcifications (yes/no, location, severity), pancreatic atrophy (yes/no, severity), pancreatic duct dilatation (yes/no, degree of dilatation), pancreatic duct stricture/obstruction (yes/no), pancreatic duct irregularities (yes/no, degree), side branch dilatation (yes/no), pancreatic/peripancreatic inflammatory changes (yes/no), pseudocyst (yes/no, number, size of largest pseudocyst), pancreatic mass (yes/no, location), common bile duct stricture (yes/no), common bile duct dilatation (yes/no), liver related findings (findings for cirrhosis, portal hypertension, splenomegaly, splenic vein thrombosis, ascites, hepatomegaly, etc.) (yes/no).
The choice of treatment for pain in chronic pancreatitis may be determined by combinations of findings on imaging studies. For example, a patient with pain who has pancreatic duct stricture with or without pancreatic duct stones is a candidate for attempting endoscopic therapy with or without lithotripsy. We therefore used a hierarchical algorithm based on physician responses to individual imaging findings on the most recent imaging studies to create the following four categories of imaging findings:
-
a)
Obstructive findings: presence of pancreatic duct obstruction/stricture irrespective of the presence of any additional imaging finding; in remaining patients,
-
b)
Inflammatory findings: presence of inflammation or inflammatory mass irrespective of the presence of any additional imaging finding; in remaining patients,
-
c)
Pseudocysts: presence of pseudocyst(s) irrespective of the presence of any additional imaging finding; and,
-
d)
All other patients: all remaining patients were assigned to this category.
Since pancreatic duct obstruction can be a mechanism for the development of pseudocysts, we also performed a sensitivity analyses by creating three categories (obstructive, inflammatory, others) where the hierarchical model incorporated pseudocysts into the obstructive category.
Comparison groups and Statistical analyses
Descriptive analyses are presented as proportions for categorical data and mean ± standard deviation (SD) or median and interquartile range (IQR) for continuous data as applicable. We evaluated the prevalence of individual imaging findings in all patients and within subgroups of pain patterns (no pain, pain subtype A-E), temporal nature of pain (intermittent, constant), and, severity of pain (mild-moderate, severe), and within each of the imaging classification categories (obstructive, inflammation, pseudocyst, others). We then evaluated the distribution of pain patterns (no pain, pain subtype A-E), temporal nature of pain (intermittent, constant), and, severity of pain (mild-moderate, severe) within each of the imaging classification categories (obstructive, inflammation, pseudocyst, others). Distribution of imaging findings and pain categories were evaluated by Chi-squared tests. We performed a sensitivity analyses by evaluating the distribution of individual imaging finding and the distribution of pain patterns in three imaging classification categories. Data analysis was performed using SAS program version 9.3. Two-tailed p-values <0.05 were considered statistically significant.
RESULTS
Demographics and etiology
Of the 521 chronic pancreatitis patients enrolled in the NAPS2-CV study, complete data on imaging studies was present in 518 patients who formed the final cohort for this analysis. Of these, 55% were male, 87.6% were white and the mean age at the time of enrollment was 52±14.6 years. The most common physician defined etiologies were - alcohol (45.8%), idiopathic (24.3%), genetic (9.5%), obstructive (7.0%) and others (13.5%).
Pain patterns
Of the 518 patients, 91 (15.6%) reported no abdominal pain from chronic pancreatitis in the year preceding enrollment in the study. The distribution of patients in different categories of pain (A-E) and by temporal nature and severity is shown in Table 1. The most common individual pain pattern was pattern D (constant mild pain plus episodes of severe pain) which was seen in about 45% patients. Only a small fraction of patients (3.7%) reported constant severe pain that does not change (i.e. pattern E). The description of pain perceived by patients could be classified as constant in about 50% patients and about two-thirds of patients characterized their pain as severe (Table 1). Also, data on the use of narcotics and frequency (intermittent vs constant) suggested increasing use of narcotics by severity and constant nature. For example, 58% of patients reporting constant pain used narcotics in a constant fashion as compared to 16% in those with intermittent pain.
Imaging findings and categories
The distribution of individual findings on imaging studies in all patients and in the different imaging categories is shown is Table 2. The most common imaging findings included – pancreatic duct dilation (67.6%), atrophy (57.3%), calcifications (55.2%) and pancreatic duct irregularity (51.4%). A history of pancreatic pseudocyst(s) at any time was seen in 32.2% and on the most recent studies in 26.4% patients. Common bile duct dilatation or stricture was observed in 17.8% and 11.8% patients, respectively and changes of either cirrhosis or portal hypertension were present in 12.4% patients.
Table 2.
Distribution of individual imaging findings in CP patients in imaging categories
| Individual imaging findings | All patients (n=518) | Imaging category | p-value | |||
|---|---|---|---|---|---|---|
| Obstructive (n=103) | Inflammatory (n=130) | Pseudocysts (n=52) | Others (n=233) | |||
| Calcification | 276 (53.3) | 63 (61.2) | 74 (56.9) | 25 (48.1) | 114 (48.9) | 0.134 |
| Calcification* | 286 (55.2) | 67 (65.0) | 76 (58.5) | 26 (50.0) | 117 (50.2) | 0.057 |
| Atrophy | 281 (54.2) | 61 (59.2) | 54 (41.5) | 35 (67.3) | 131 (56.2) | 0.004 |
| Atrophy* | 297 (57.3) | 64 (62.1) | 57 (43.8) | 36 (69.2) | 140 (60.1) | 0.002 |
| PD Dilatation | 317 (61.2) | 90 (87.4) | 66 (50.8) | 37 (71.2) | 124 (53.2) | <0.001 |
| PD Dilatation* | 350 (67.6) | 94 (91.3) | 73 (56.2) | 41 (78.8) | 142 (60.9) | <0.001 |
| PD Stricture / Obstruction | 103 (19.9) | 103 (100) | 0 (0) | 0 (0) | 0 (0) | - |
| PD Stricture / Obstruction* | 142 (27.4) | 103 (100) | 17 (13.1) | 4 (7.7) | 18 (7.7) | - |
| Pseudocyst | 137 (26.4) | 25 (24.3) | 60 (46.2) | 52 (100) | 0 (0) | <0.001 |
| Pseudocyst* | 167 (32.2) | 35 (34.0) | 68 (52.3) | 52 (100) | 12 (5.2) | <0.001 |
| PD Irregularity | 266 (51.4) | 87 (84.5) | 55 (42.3) | 29 (55.8) | 95 (40.8) | <0.001 |
| Inflammation | 176 (34.0) | 46 (44.7) | 130 (100) | 0 (0) | 0 (0) | - |
| Pancreatic Mass | 30 (5.8) | 10 (9.7) | 11 (8.5) | 0 (0) | 9 (3.9) | 0.025 |
| Abnormal Side Branches | 213 (41.1) | 57 (55.3) | 44 (33.8) | 26 (50.0) | 86 (36.9) | 0.002 |
| CBD Stricture | 61 (11.8) | 21 (20.4) | 17 (13.1) | 6 (11.5) | 17 (7.3) | 0.007 |
| CBD Dilatation | 92 (17.8) | 29 (28.2) | 23 (17.7) | 9 (17.3) | 31 (13.3) | 0.012 |
| Liver related findings | 64 (12.4) | 16 (15.5) | 14 (10.8) | 10 (19.2) | 24 (10.3) | 0.22 |
Findings are those reported on the most recent imaging studies, except for those marked with an (*) which reflects findings reported on imaging studies performed anytime
Imaging findings were categorized as obstructive in 19.9%, inflammatory in 25.1% and pseudocysts in 10% of patients; findings in the remaining patients did not fit into any of the three categories (45%). As expected, the prevalence of individual findings differed between imaging categories. Patients with obstructive findings were more likely to have pancreatic duct dilatation, pancreatic duct irregularity, abnormal side branches, common bile duct stricture and dilatation and a tendency towards higher prevalence of calcifications. Patients with inflammatory findings were less likely to have atrophy and higher prevalence of pseudocysts. Patients with pseudocysts were also more likely to have pancreatic ductal dilatation and abnormal side branches.
Association of pain patterns with imaging findings
The distribution of individual imaging findings was similar in patients with different pain patterns (no pain, A-E pain pattern), temporal nature of pain (no pain, intermittent, constant) or pain severity (no pain, mild-moderate, severe) (Tables 3 and 4).
Table 3.
Individual imaging findings by pain pattern in chronic pancreatitis patients in the NAPS2-CV study
| Variable | All Patients (518) | Pain pattern | P-Value | |||||
|---|---|---|---|---|---|---|---|---|
| No Pain (81) | A (67) | B (23) | C (98) | D (229) | E (19) | |||
| Calcification | 276 (53.3) | 49 (60.5) | 37 (55.2) | 12 (50) | 58 (59.2) | 107 (46.7) | 13 (68.4) | 0.11 |
| Calcification* | 286 (55.2) | 51 (63) | 41 (61.2) | 12 (50) | 59 (60.2) | 109 (47.6) | 14 (73.7) | 0.0324 |
| Atrophy | 281 (54.3) | 50 (61.7) | 37 (55.2) | 16 (66.7) | 48 (49) | 121 (52.8) | 9 (47.4) | 0.41 |
| Atrophy* | 297 (57.3) | 53 (65.4) | 38 (56.7) | 18 (75) | 50 (51) | 128 (55.9) | 10 (52.6) | 0.21 |
| PD Dilatation | 317 (61.2) | 50 (61.7) | 44 (65.7) | 14 (58.3) | 64 (65.3) | 135 (59) | 10 (52.6) | 0.79 |
| PD Dilatation* | 350 (67.6) | 52 (64.2) | 49 (73.1) | 15 (62.5) | 70 (71.4) | 151 (65.9) | 13 (68.4) | 0.76 |
| PD Stricture/Obstruction | 103 (19.9) | 10 (12.4) | 16 (23.9) | 2 (8.3) | 26 (26.5) | 46 (20.1) | 3 (15.8) | 0.13 |
| PD Stricture/Obstruction* | 142 (27.4) | 17 (21) | 22 (32.8) | 3 (12.5) | 30 (30.6) | 63 (27.5) | 7 (36.8) | 0.24 |
| Pseudocyst | 137 (26.5) | 16 (19.8) | 24 (35.8) | 5 (20.8) | 18 (18.4) | 70 (30.6) | 4 (21.1) | 0.0544 |
| Pseudocyst* | 167 (32.2) | 19 (23.5) | 28 (41.8) | 5 (20.8) | 24 (24.5) | 86 (37.6) | 5 (26.3) | 0.0229 |
| PD Irregularity | 266 (51.4) | 34 (42) | 38 (56.7) | 14 (58.3) | 57 (58.2) | 115 (50.2) | 8 (42.1) | 0.25 |
| Inflammation | 176 (34) | 23 (28.4) | 19 (28.4) | 4 (16.7) | 37 (37.8) | 82 (35.8) | 11 (57.9) | 0.0497 |
| Pancreatic Mass | 30 (5.8) | 6 (7.4) | 3 (4.5) | 1 (4.2) | 9 (9.2) | 9 (3.9) | 2 (10.5) | 0.31 |
| Abnormal Side Branches | 213 (41.1) | 27 (33.3) | 35 (52.2) | 11 (45.8) | 36 (36.7) | 97 (42.4) | 7 (36.8) | 0.24 |
| CBD Stricture | 61 (11.8) | 9 (11.1) | 10 (14.9) | 3 (12.5) | 19 (19.4) | 19 (8.3) | 1 (5.3) | 0.0911 |
| CBD Dilatation | 92 (17.8) | 16 (19.8) | 17 (25.4) | 4 (16.7) | 20 (20.4) | 32 (14) | 3 (15.8) | 0.31 |
| Changes of Cirrhosis and/or Portal Hypertension | 64 (12.4) | 7 (8.6) | 9 (13.4) | 2 (8.3) | 11 (11.2) | 30 (13.1) | 5 (26.3) | 0.40 |
Findings are those reported on the most recent imaging studies, except for those marked with an (*) which reflects findings reported on imaging studies performed anytime
Table 4.
Individual imaging findings by temporal nature and severity of pain in chronic pancreatitis patients in the NAPS2-CV study
| Variable | All Patients (518) |
Temporal | Severity | ||||||
|---|---|---|---|---|---|---|---|---|---|
| No pain (81) |
Intermittent (165) |
Constant (272) |
p- value |
No pain (81) |
Mild (91) |
Severe (346) |
p- value |
||
| Calcification | 276 (53.3) | 49 (60.5) | 95 (57.6) | 132 (48.5) | 0.07 | 49 (60.5) | 49 (53.9) | 178 (51.5) | 0.34 |
| Calcification* | 286 (55.2) | 51 (63) | 100 (60.6) | 135 (49.6) | 0.0256 | 51 (63) | 53 (58.2) | 182 (52.6) | 0.20 |
| Atrophy | 281 (54.3) | 50 (61.7) | 85 (51.5) | 146 (53.7) | 0.31 | 50 (61.7) | 53 (58.2) | 178 (51.5) | 0.17 |
| Atrophy* | 297 (57.3) | 53 (65.4) | 88 (53.3) | 156 (57.4) | 0.20 | 53 (65.4) | 56 (61.5) | 188 (54.3) | 0.13 |
| PD Dilatation | 317 (61.2) | 50 (61.7) | 108 (65.5) | 159 (58.5) | 0.34 | 50 (61.7) | 58 (63.7) | 209 (60.4) | 0.84 |
| PD Dilatation* | 350 (67.6) | 52 (64.2) | 119 (72.1) | 179 (65.8) | 0.31 | 52 (64.2) | 64 (70.3) | 234 (67.6) | 0.69 |
| PD Stricture/Obstruction | 103 (19.9) | 10 (12.4) | 42 (25.5) | 51 (18.8) | 0.0424 | 10 (12.4) | 18 (19.8) | 75 (21.7) | 0.17 |
| PD Stricture/Obstruction* | 142 (27.4) | 17 (21) | 52 (31.5) | 73 (26.8) | 0.21 | 17 (21) | 25 (27.5) | 100 (28.9) | 0.36 |
| Pseudocyst | 137 (26.5) | 16 (19.8) | 42 (25.5) | 79 (29) | 0.24 | 16 (19.8) | 29 (31.9) | 92 (26.6) | 0.20 |
| Pseudocyst* | 167 (32.2) | 19 (23.5) | 52 (31.5) | 96 (35.3) | 0.13 | 19 (23.5) | 33 (36.3) | 115 (33.2) | 0.16 |
| PD Irregularity | 266 (51.4) | 34 (42) | 95 (57.6) | 137 (50.4) | 0.06 | 34 (42) | 52 (57.1) | 180 (52) | 0.13 |
| Inflammation | 176 (34) | 23 (28.4) | 56 (33.9) | 97 (35.7) | 0.48 | 23 (28.4) | 23 (25.3) | 130 (37.6) | 0.0453 |
| Pancreatic Mass | 30 (5.8) | 6 (7.4) | 12 (7.3) | 12 (4.4) | 0.37 | 6 (7.4) | 4 (4.4) | 20 (5.8) | 0.70 |
| Abnormal Side Branches | 213 (41.1) | 27 (33.3) | 71 (43) | 115 (42.3) | 0.30 | 27 (33.3) | 46 (50.6) | 140 (40.5) | 0.0661 |
| CBD Stricture | 61 (11.8) | 9 (11.1) | 29 (17.6) | 23 (8.5) | 0.02 | 9 (11.1) | 13 (14.3) | 39 (11.3) | 0.72 |
| CBD Dilatation | 92 (17.8) | 16 (19.8) | 37 (22.4) | 39 (14.3) | 0.09 | 16 (19.8) | 21 (23.1) | 55 (15.9) | 0.25 |
| Changes suggestive of Cirrhosis and/or Portal Hypertension | 64 (12.4) | 7 (8.6) | 20 (12.1) | 37 (13.6) | 0.49 | 7 (8.6) | 11 (12.1) | 46 (13.3) | 0.52 |
Findings are those reported on the most recent imaging studies, except for those marked with an (*) which reflects findings reported on imaging studies performed anytime
Three representative patients are demonstrated in the Figure 1, 2, 3.
Figure 1.
Abdominal CT scan demonstrates dense calcification of the pancreatic head with upstream ductal dilation and atrophy. This 51-year-old male with chronic pancreatitis due to alcohol described his pain as constant, mild to moderate intensity, with episodes of more severe pain.
Figure 2.
Abdominal CT scan shows calcific disease in the pancreatic head (a) associated with upstream ductal dilation and a stone at the neck (b). He describes his pain as usually pain free but with episodes of mild to moderate pain.
Figure 3.
Abdominal CT shows calcifications throughout the entire pancreas most significant at the tail associated with mild to moderate ductal dilatation and atrophy. This 51-year-old male with chronic pancreatitis due to alcohol denied abdominal pain or any episodes of pain.
Association of pain patterns with imaging categories
The distribution of pain patterns in the different imaging categories is shown in Tables 5 and 6. Although statistical differences were noted in the distribution of imaging findings with pain patterns, they do not appear to be clinically relevant. In 39 patients, findings of pancreatic duct dilatation/stricture were seen on prior imaging but not on the most recent imaging studies. Analysis of the distribution of the pain patterns in the different imaging categories was more significant but similar after exclusion of these patients from the analyses.
Table 5.
Distribution of temporal pain patterns based on temporal nature in patients with different imaging categories
| Imaging Category | Temporal Nature of pain | |||
|---|---|---|---|---|
| No pain (n=81) | Intermittent (n=165) | Constant (n=272) | p-value | |
| Obstructive (n=103) | 10 (12.4) | 42 (25.5) | 51 (18.6) | 0.0891 |
| Inflammatory (n=130) | 18 (22.2) | 40 (24.2) | 72 (26.5) | |
| Pseudocysts (n=52) | 6 (7.4) | 19 (11.5) | 27 (9.9) | |
| Others (n=233) | 47 (58.0) | 64 (38.8) | 122 (44.9) | |
Analysis after removing 39 patients with pancreatic duct stricture/dilatation from inflammation, pseudocyst and other categories showed similar results
Table 6.
Distribution of pain patterns based on severity in patients with different imaging categories
| Imaging Category | Severity | |||
|---|---|---|---|---|
| No pain (n=81) | Mild (n=91) | Severe (n=346) | p-value | |
| Obstructive (n=103) | 10 (12.4) | 18 (19.8) | 75 (21.7) | 0.0405 |
| Inflammatory (n=130) | 18 (22.2) | 18 (19.8) | 94 (27.1) | |
| Pseudocysts (n=52) | 6 (7.4) | 15 (16.5) | 31 (9.0) | |
| Others (n=233) | 47 (58.0) | 40 (44.0) | 146 (42.2) | |
Analysis after removing 39 patients with pancreatic duct stricture/dilatation from inflammation, pseudocyst and other categories showed similar results
Sensitivity analyses
On univariate analyses, increasing age, never/past smoker status and the amount of smoking were associated with a higher likelihood of having no pain. Due to this, we repeated the analyses by stratification by age groups (<35, 35-65 and 65+ years at the time of enrollment, sex, smoking status) and with alcohol etiology (yes, no). These analyses showed generally similar results and some comparisons had small sample sizes limiting the ability to make meaningful interpretations (data not shown). We also evaluated the distribution of pain patterns in three imaging categories (obstructive, inflammatory, others) and found generally similar results (Tables 3-6).
DISCUSSION
In this large multicenter cross-sectional study, we found that patients with chronic pancreatitis had variable levels and types of abdominal pain but the presence and type of pain had little correlation with pancreatic morphology on cross-sectional radiographic or endoscopic imaging. Indeed, patients with obstructive anatomy were as likely to have no pain as to have severe daily pain. Patients with inflammatory disease had a trend toward having abdominal pain, often severe.
Our findings are consistent with prior prospective (3-5, 9) and retrospective (6-8) observations that have shown poor correlation between pain severity and abdominal imaging characteristics but provide additional and important information on imaging and pain pattern, which is a major determinant of pain impact (1).
A variety of radiographic and endoscopic imaging modalities were used to determine the severity of chronic pancreatitis in the previous studies, including ductal dilatation at pancreatography, fluid collections and calcifications on cross-sectional abdominal imaging, a number of findings on endoscopic ultrasonography, as well as more recently using diffusion weighted imaging by magnetic resonance imaging. In a small study of 54 patients referred for differential nerve block, no difference in rates of pain were found based on severity of abdominal imaging using a normal, mild and severe category (8). In a study examining pancreatography pre and post pancreatic ductal stenting (6), follow-up after stenting demonstrated variable effects on duct caliber but these were unrelated to a patient's clinical response again suggesting the absence of a morphologic-physiologic correlation. Using MRI diffusion-weighted imaging, which provides a measure of fibrotic pancreatic changes, 23 patients with chronic pancreatitis and pain were compared to controls (9). In this study, differences were found between the two groups but atrophy and duct related findings were not associated with pain or other symptoms. Thus, regardless of the modality utilized, no apparent correlation between pattern and severity of pain has been identified. Using EUS, Gardner and colleagues (7) showed that patients presenting with steatorrhea with or without pain had more parenchymal changes than those presenting with pain alone although not statistically different. When combined with ductal changes, those with pain had less endosonographic abnormalities as compared to steatorrhea with or without pain. The data was not examined to evaluate pain only versus patients without pain (7).
The mechanism(s) underlying this disparity between morphology and pain remains speculative. It is well established that the source of abdominal pain in patients with chronic pancreatitis is complex and, as noted above, may be multifactorial. Not only can there be complications of disease including inflammation, obstruction (bile duct, pancreatic duct) as well as local complications including pseudocysts, local and central neuropathic changes also participate (2). Such neuropathic changes include central sensitization, impaired inhibitory pain modulation, as well as abnormalities of central pain processing (14-16). Given these changes, it may not be surprising then that no correlation between the imaging and pain pattern was identified. In patients with longstanding pain associated with chronic pancreatitis, alterations in cerebral cortical thickness have been described further supporting central mechanisms for pain (17). If the duration of pain is important in leading to central mechanisms rather than peripheral (pancreatic) causes of pain, then it would be important for studies to specifically address the duration of abdominal pain. Neither our study nor the few others evaluating this morphologic symptomatic correlative have obtained such information. Until better and more widely available methods are available to study these disparate mechanisms, it may be difficult to assign specific anatomic and neurologic mechanisms to the cause of pain as well as radiographic findings.
Several limitations of our study deserve mention. Firstly, all imaging studies were not reviewed at a single site by experienced gastrointestinal radiologists. Nevertheless, patients were recruited from academic centers commonly dealing with such patients. In addition, such a multicenter collection of patients would support generalizability of the results. Secondly, reports of pain were obtained from a questionnaire and thus are subject to patient bias given the absence of any objective measure of pain as well as recall bias. However, the distribution of pain patterns in the NAPS2-CV study is generally similar to the original NAPS2 cohort (1) thereby validating the use of these pain patterns (A-E, temporal nature and severity). Thirdly, a large fraction of patients (~45%) were classified into the “other” imaging category. It is possible that our unbiased post hoc classification of patients into clinically relevant imaging categories based on a combination of individual imaging findings underestimated patients in other categories (obstructive, inflammatory, pseudocyst) or resulted in misclassification. It is difficult to know if or how this approach would have differed from asking the enrolling physician to interpret the available imaging and classify patients into such categories. Future studies should consider using predefined criteria to classify patients into specific categories for correlation with pain patterns. Although we asked patients about their pain experiences in the year before study entry, we did not systematically capture information on the pain symptoms at the time of imaging studies. Finally, the aim of our analysis was to evaluate whether patterns of pain are related to specific imaging findings that can potentially inform their mechanism (e.g. obstruction, inflammation, etc.). Our study was not designed to assess whether the pattern of pain has any bearing on the efficacy of interventions or treatment. This question should be addressed in future investigations.
In summary, our large study supports prior observations of an absence between pancreatic morphology assessed radiographically and presence and severity of abdominal pain. These findings have implications for patient management and pathogenesis of pain in these patients.
Acknowledgments
Funding:
The study was supported by R01DK061451 (DCW), R01 DK077906 (DY) and UL1 RR024153 from the National Center for Research Resources (NCRR), a component of the National Institutes of Health (NIH), and NIH Roadmap for Medical Research.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Disclosures: none
The study was IRB approved by each individual center.
Author Contributions:
Study design: CMW, DCW, MAA, SRW, DY
Writing: CMW, DY
Ascertained and phenotyped patients: SA, MAA, PAB, REB, GC, DLC, CEF, NG, TBG, AG, ML, TM, JR, BS, SS, AS, CMW, DY,
Statistical Analysis: SRW, DY, TY
All authors reviewed and approved the manuscript.
REFERENCES
- 1.Mullady DK, Yadav D, Amann ST, O'Connell MR, Barmada MM, Elta GH, Scheiman JM, Wamsteker EJ, Chey WD, et al. Type of pain, pain-associated complications, quality of life, disability and resource utilization in chronic pancreatitis: a prospective cohort study. Gut. 2011;60:77–84. doi: 10.1136/gut.2010.213835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Poulsen JL, Olesen SS, Malver LP, Frøkjær, Drewes AM. Pain and chronic pancreatitis: A complex interplay of multiple mechanisms. World J Gastroenterol. 2013;19:7282–7291. doi: 10.3748/wjg.v19.i42.7282. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Bornman PC, Marks IN, Girdwood AH, Clain JE, Narunsky L, Clain DJ, Wright JP. Is pancreatic duct obstruction or stricture or major cause of pain in calcific pancreatitis? Br J Surg 2980. 67:425–8. doi: 10.1002/bjs.1800670614. [DOI] [PubMed] [Google Scholar]
- 4.Jensen AR, Matzen P, Malchow-Møller A, Christoffersen I. Pattern of pain, duct morphology, and pancreatic function in chronic pancreatitis. A comparative study. Scand J Gastroenterol. 1984;19:334–8. [PubMed] [Google Scholar]
- 5.Malfertheiner P, Büchler M, Stanescu A, Ditschuneit H. Pancreatic morphology and function in relationship to pain in chronic pancreatitis. Int J Pancreatol. 1987;2:59–66. doi: 10.1007/BF02788349. [DOI] [PubMed] [Google Scholar]
- 6.Morgan DE, Smith JK, Hawkins K, Wilcox CM. Endoscopic stent therapy in advanced chronic pancreatitis: relationships between ductal changes, clinical response, and stent patency. Am J Gastroenterol. 2003;98:821–6. doi: 10.1111/j.1572-0241.2003.07381.x. [DOI] [PubMed] [Google Scholar]
- 7.Gardner TB, Janec EM, Gordon SR. Relationship between patients symptoms and endosonographic findings in chronic pancreatitis. Pancreatology. 2009;9:398–403. doi: 10.1159/000181178. [DOI] [PubMed] [Google Scholar]
- 8.Bahuva R, Walsh RM, Kapural L, Stevens T. Morphologic abnormalities are poorly predictive of visceral pain in chronic pancreatitis. Pancreas. 2013;42:6–10. doi: 10.1097/MPA.0b013e318258cd9c. [DOI] [PubMed] [Google Scholar]
- 9.Frøkjaer JB, Olesen SS, Drewes AM. Fibrosis, atrophy, and ductal pathology in chronic pancreatitis are associated with pancreatic function but independent of symptoms. Pancreas. 2013;42:1182–1187. doi: 10.1097/MPA.0b013e31829628f4. [DOI] [PubMed] [Google Scholar]
- 10.Whitcomb DC, Yadav D, Adam S, et al. Multicenter approach to recurrent acute and chronic pancreatitis in the United States: the North American Pancreatitis Study 2 (NAPS2). Pancreatology. 2008;8(4-5):520–31. doi: 10.1159/000152001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Etemad B, Whitcomb DC. Chronic pancreatitis: diagnosis, classification, and new genetic developments. Gastroenterology. Feb. 2001;120(3):682–707. doi: 10.1053/gast.2001.22586. [DOI] [PubMed] [Google Scholar]
- 12.Melzack R. The short-form McGill Pain Questionnaire. Pain. 1987;30:191–197. doi: 10.1016/0304-3959(87)91074-8. [DOI] [PubMed] [Google Scholar]
- 13.Strand LI, Ljunggren AE, Bogen B, Ask T, Johnsen TB. The short-form McGill Pain Questionnaire as an outcome measure: test-retest reliability and responsiveness to change. Eur J Pain. 2008;12:917–925. doi: 10.1016/j.ejpain.2007.12.013. [DOI] [PubMed] [Google Scholar]
- 14.Rizk MK, Tolba R, Kapural L, Mitchell J, Lopez R, Mahboobi R, Vrooman B, Mekhail N. Differential epidural block predicts the success of visceral block in patients with chronic visceral abdominal pain. Pain Pract. 2012;12:595–601. doi: 10.1111/j.1533-2500.2012.00548.x. [DOI] [PubMed] [Google Scholar]
- 15.Dimcevski G, Sami SA, Funch-Jensen P, Le Pera D, Valeriani M, Arendt-Nielsen L, Drewes AM. Pain in chronic pancreatitis: the role of reorganization in the central nervous system. Gastroenterology. 2007;132:1546–56. doi: 10.1053/j.gastro.2007.01.037. [DOI] [PubMed] [Google Scholar]
- 16.Bouwense SA, Ahmed AU, Ten Broek RP, Issa Y, van Eijk CH, Wilder-Smith OH, van Goor H. Altered central pain processing after pancreatic surgery for chronic pancreatitis. Br J Surg. 2013;100:1797–804. doi: 10.1002/bjs.9322. [DOI] [PubMed] [Google Scholar]
- 17.Frøkjær JB, Bouwense SA, Olesen SS, Lundager FH, Eskildsen SF, van Goor H, Wilder-Smith OH, Drewes AM. Reduced cortical thickness of brain areas involved in pain processing in patients with chronic pancreatitis. Clin Gastroenterol Hepatol. 2012;10:434–8. doi: 10.1016/j.cgh.2011.11.024. [DOI] [PubMed] [Google Scholar]