NDT Advance Access published online on December 9, 2007
Nephrology Dialysis Transplantation, doi:10.1093/ndt/gfm769
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Frequency and Severity of Acute Pancreatitis in Chronic Dialysis Patients
1 Clinic for General Internal Medicine, Medical Centre, Municipal Clinic of Lüneburg, Lüneburg, Germany 2 European Institute of Oncology, Milan, Italy 3 Medical College, Valhalla, NY, USA
Correspondence and offprint requests to: P. G. Lankisch, Reiherstieg 23, D-21337 Lüneburg, Germany. Tel: +49-4131-403503; +49-4131-249495; E-mail: paulgeorg.lankisch{at}t-online.de
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and methodsStatistical methodsResultsDiscussionConclusionsReferences |
|---|
Background. The incidence and severity of acute pancreatitis in patients undergoing dialysis treatment are unknown.
Methods. A questionnaire asking for the incidence and the severity of a first attack of acute pancreatitis in chronic dialysis patients in the year 2002 was sent to the members of QuaSi-Niere gGmbH, an organization representing almost all dialysis centres in Germany. A second questionnaire was sent to those who reported such patients.
Results. Response rates for the first and second questionnaire were 72% (832 out of 1150 centres) and 100% (72 out of 72 centres), respectively. After the exclusion of patients with invalid data, 55 patients with acute pancreatitis remained: 46 patients out of 68 715 haemodialysis (HD) patients (incidence rate 67/100 000/year; 95% confidence interval, 49 to 89/100 000/year) and 9 out of 3386 peritoneal dialysis (PD) patients (incidence rate 266/100 000/year; 95% confidence interval, 122 to 504/100 000/year; Fisher's exact test: P = 0.002). Twenty-eight patients (51%) had a known risk factor for acute pancreatitis. When these were excluded, the incidence of pancreatitis of unknown aetiology was 32/100 000/year (20–48) for HD patients (n, 22) and 148/100 000/year (48–345) for PD patients (n, 3; Fisher's exact test: P = 0.016). PD patients required hospital admission more frequently than HD patients (100% versus 76%) and suffered more frequently from necrotizing pancreatitis (50% versus 19%).
Conclusions. Dialysis—especially PD—is another risk factor that increases the susceptibility of the pancreas to acute pancreatitis. Acute pancreatitis in patients undergoing PD is more frequent and seems to be more severe than in those receiving HD treatment.
Keywords: acute pancreatitis; dialysis; haemodialysis; peritoneal dialysis
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and methodsStatistical methodsResultsDiscussionConclusionsReferences |
|---|
Several postmortem studies have shown that the pancreas is affected by uraemia [1–3]. Some case reports and small studies suggest that the risk of acute pancreatitis is increased in patients with end-stage renal disease undergoing peritoneal dialysis (PD) as compared to those undergoing haemodialysis (HD) treatment [4–9], whereas others could not confirm this connection [10].
Recently, two interesting studies on this subject have been published. In one retrospective single-centre cohort study from the Netherlands, patients on long-term PD (but not HD) were found to have a higher incidence of acute pancreatitis than the general population [11]. Another retrospective case-control study from the United States reported that PD is a risk factor for acute pancreatitis in the black population and that there are no statistical differences in acute pancreatitis-related mortality and morbidity between both dialysis procedures [12].
Since the number of dialysis patients waiting for kidney transplants is increasing, the conclusions of both studies have a substantial clinical impact on the care of these patients and were therefore questioned. The aim of our study was to determine the incidence and severity of acute pancreatitis in a substantially larger patient population.
|
Patients and methods |
|---|
|
TopAbstractIntroductionPatients and methodsStatistical methodsResultsDiscussionConclusionsReferences |
|---|
In the year 2003, we sent out a questionnaire to the members of the QuaSi-Niere gGmbH asking for the incidence and the severity of a first attack of acute pancreatitis in chronic dialysis patients in 2002. This organization represents almost all dialysis centres in Germany (n, 1150; response rate: 832 or 72%). There were no obvious differences between reporting and non-reporting centres concerning the size and the geographical distribution of the centres. Chronic dialysis was defined as a total dialysis period longer than 6 weeks [11].
Acute pancreatitis is usually defined as acute abdominal pain with an increase in serum levels of pancreatic enzymes. Serum amylase and lipase are frequently elevated in chronic renal failure and may exceed three times the upper limit of normal even in the absence of acute pancreatitis [13]. Since the severity of acute pancreatitis is not dependent on the degree of pancreatic enzyme elevation, a severe attack may occur in a substantial number of patients where the enzyme elevation is less than three times the upper limit of normal [14]. In our study, the diagnosis of acute pancreatitis was made when, in addition to acute abdominal pain, enzyme elevation was present. When this elevation was less than three times the upper limit of normal, the diagnosis had to be supported by an abnormal imaging procedure [ultrasound (US) or contrast-enhanced computed tomography (CT); see Table 3].
|
Alcohol abuse was assumed to be the cause of acute pancreatitis in patients reporting >60 g pure alcohol/day as was biliary disease in patients who demonstrated stones in the gallbladder or common bile duct upon different imaging procedures. Many patients (26/55; in HD patients 19 out of 27 and in PD patients 7 out of 9; differences not significant) received drugs that, in rare cases, have been thought to cause acute pancreatitis [15]. Because none of these patients discontinued this medication after the onset of acute pancreatitis and no relapse of pancreatitis occurred, however, drug-induced pancreatitis seemed unlikely here. Other factors were considered to be responsible in patients who had undergone interventional or surgical procedures, suffered from immunological disorders, experienced dialysis-related complications, or had systemic shock, hypercalcaemia (>12 mg/dl), hypertriglyceridaemia (>500 mg/dl) or posttraumatic pancreatitis. In the remaining patients, acute idiopathic pancreatitis was diagnosed.
Acute pancreatitis was staged as either mild (abdominal pain, elevated pancreatic enzymes; no imaging procedures deemed necessary by the members of the centre), moderate (mild plus an abnormal US or CT, no necrosis) or severe (moderate plus evidence of necrosis on contrast-enhanced CT).
Centres which reported patients with acute pancreatitis (72 centres; 77 HD and 17 PD patients) received a second questionnaire and were asked to send in anonymously the medical charts of these patients for re-evaluation of the diagnosis with consideration of age, gender, aetiology, medication, onset of symptoms (during dialysis/independent of the procedure), initiation of dialysis, pain intensity, signs of peritonitis, amylase and lipase elevation, results of imaging procedures, necessity for hospital admission, relapse of acute pancreatitis and mortality. Altogether, 31 HD and 8 PD patients with initially reported acute pancreatitis were excluded for the following reasons: occurrence of acute pancreatitis in a year other than 2002 (6 HD patients, 1 PD patient), occurrence prior to the initiation of dialysis (5; 0) or the misdiagnosis of an acute attack of already proven chronic pancreatitis (13; 3). In 11 patients, the diagnosis of acute pancreatitis could not be verified on the basis of medical reports (7; 4).
|
Statistical methods |
|---|
|
TopAbstractIntroductionPatients and methodsStatistical methodsResultsDiscussionConclusionsReferences |
|---|
The incidence rate of acute pancreatitis during dialysis was calculated dividing the number of events during the year 2002 by the number of patients at risk. Ninety-five percent confidence intervals were calculated using the Fisher's exact test. Comparison of the rate of pancreatitis in patients receiving PD and HD was assessed by the Fisher's exact binomial test. The Wilcoxon two-sample test was used to compare the time interval from first dialysis to the development of pancreatitis and the Fisher's exact test to compare the proportion of other patients’ characteristics in the two dialysis groups. All P-values were two-sided.
This study was approved by the Ethical Committee of the Georg-August-University of Göttingen, no. 25/1/03.
|
Results |
|---|
|
TopAbstractIntroductionPatients and methodsStatistical methodsResultsDiscussionConclusionsReferences |
|---|
The incidence of first-time acute pancreatitis in these dialysis patients was 75/100 000/year for both procedures, 67/100 000/year for HD and 266/100 000/year for PD.
Altogether, 46 cases of acute pancreatitis were found in 45 centres out of 68 715 HD patients (incidence rate 67/100 000/year, 95% confidence interval, 49 to 89/100 000/year) and in 9 patients with acute pancreatitis out of 3386 PD patients [incidence rate 266/100 000/year, 95% confidence interval, 122 to 504/100 000/year, Fisher's exact test: P = 0.002 (Table 1)].
|
The age of the patients did not differ significantly between the groups [59.5 years in HD patients and 59.0 years in PD patients (median values; P = 0.90)]. There were also no significant differences in ages among the aetiological subgroups.
The two most common aetiological factors were alcohol abuse in 7 HD patients and 1 PD patient, and biliary disease in 14 HD patients and 3 PD patients. Acute pancreatitis was related to other defined causes in three HD patients (immune vasculitis, haemolysis after HD and prior coronary bypass surgery). When these patients with a known cause of acute pancreatitis [28 (51%) out of 55 patients] were excluded, the incidence rate was 32/100 000/year, 95% confidence interval, 20 to 48/100 000/year in HD patients and 148/100 000/year, 95% confidence interval, 48 to 345/100 000/year in PD patients, P = 0.016 (Table 1).
There was a minimal difference in the time interval between first dialysis and onset of acute pancreatitis: 21 months in HD patients and 19 months in PD patients (median values; P = 0.78). No differences concerning this time interval existed among the different aetiological subgroups. Finally, no differences were found between PD and HD patients regarding gender, onset of symptoms, intensity of pain, signs of peritonitis, pancreatic enzyme elevation, abnormal imaging procedures, necessity for hospitalization and mortality (Table 2).
|
Acute pancreatitis during dialysis treatment occurred more frequently in PD patients than in HD patients: 5 (56%) out of 9 PD patients and 7 (15%) out of 46 HD patients. Among patients who had a CT within 72 h from admission, necrotizing pancreatitis was also more frequent in PD patients than in HD patients: 4 (50%) out of 8 PD patients and 5 (19%) out of 27 HD patients (Table 2).
In general, acute pancreatitis was mild or moderate in the majority of patients (84%) and severe only in 16% (Table 3). None of the patients died of acute pancreatitis.
Follow-up of the patients with acute pancreatitis showed that only 5 (9%) of the 55 patients had a second attack and that they were all from the HD group. Of these, one patient had continued alcohol abuse, another patient had a second relapse of acute pancreatitis due to biliary disease after refusing cholecystectomy and the patient whose initial acute pancreatitis had been due to immune vasculitis had several other attacks including one also after successful kidney transplantation. Furthermore, the two patients with idiopathic pancreatitis had another attack of the disease. The dialysis procedure had not been changed in these cases.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and methodsStatistical methodsResultsDiscussionConclusionsReferences |
|---|
The incidence of acute pancreatitis among the general population of Germany is 19.7/100 000/year [16]. The number of incidences calculated in our study among dialysis patients for both procedures (HD and especially PD) is considerably higher. This stands in contrast to the study from the Netherlands that reported that the risk of acute pancreatitis in patients on PD—but not on HD—was higher than in the general population.
Since our study (and probably the Dutch study as well [11]) was performed in Caucasian patients, we cannot confirm the observation of the American group that acute pancreatitis in patients undergoing chronic PD occurs only in the black population [12].
The proportion of patients with identified risk factors for acute pancreatitis was higher in our study (55%) as compared to the American study (29%) [12]. In the Dutch study, no known risk factors were reported in the seven patients with acute pancreatitis under PD [11].
The median age at the first attack of acute pancreatitis in dialysis patients did not differ from the median age at the first attack of acute pancreatitis in all patients at our hospital, which is 58 years.
Most of the patients in our study had mild to moderate pancreatitis (Table 3) when using clinical parameters and imaging procedures for staging. There were no significant differences between the groups concerning the severity of abdominal pain, signs of peritonitis and the degree of pancreatic enzyme elevation. However, some parameters showed that acute pancreatitis in the PD group was more severe than in the HD group: all patients of the first but only 67% of the second group required hospitalization for the treatment of the disease, and necrotizing pancreatitis occurred in 50% of the first and in only 19% of the second group. However, the disease had no impact on patients’ survival. In contrast, all patients with acute pancreatitis in the American study in whom a contrast-enhanced CT was performed had interstitial pancreatitis and thus the mild form of the disease [12]. In the Dutch study, however, two out of seven patients had necrotizing pancreatitis [11]. Similar to our study, mortality was low in the two other reports: none of the American patients and only one of the Dutch patients died [11,12]. In our study, the relapse rate of acute pancreatitis in patients on dialysis was 9%. The two other reports do not have any comparative data.
Our study has two disadvantages: first, its design, which is retrospective and not prospective. Second, the contrast-enhanced CT, the gold standard for morphological evaluation of the pancreas, was performed only in 23 (50%) patients. However, this procedure is frequently impossible if there is renal insufficiency. Third, there may be a possible bias due to the over- or under-reporting of acute pancreatitis. As the usual response rate for medical questionnaires does not exceed 58% [17], our response rates of 72% for the first and 100% for the second questionnaire may be considered fairly representative. However, 28% of the centres did not respond to our first questionnaire, which renders our results less representative than we would have desired. As it is more troublesome to search archives for the medical records of patients with acute pancreatitis than simply report that no such patient was seen, the incidence of acute pancreatitis in chronic dialysis patients may in truth be even higher than was reflected in our study.
Idiopathic pancreatitis in this special population may possibly be due to the cumulative effect of separate pathogenic factors: the risk of the general population; the risk related to renal insufficiency, uraemia, secondary hyperparathyroidism with hypercalcaemia, hypertriglyceridaemia and drug abuse [11] plus the risk of dialysis itself. In the group of patients with alcohol abuse and biliary disease, dialysis may render the pancreas especially susceptible to pancreatitis.
In both groups, the time between the first dialysis procedure and the first attack of acute pancreatitis corresponded. However, the occurrence of acute pancreatitis during the dialysis procedure, which was much higher in the PD group than in the HD group, suggests that within the dialysis process itself there yet exist unidentified risk factors. These may be the composition and amount of fluid given and dialysate through the peritoneum of the lesser sac to the anterior surface of the pancreas, causing chemical irritation. Finally, another potential mechanism could be linked to the calcium in the peritoneal dialysate. Calcium could diffuse through the peritoneum, causing local hypercalcaemia of the pancreas, even though the systemic calcium levels may not be elevated [12].
In summary, our nationwide study including >70 000 dialysis patients establishes that dialysis—especially PD—is another risk factor that increases the susceptibility of the pancreas to acute pancreatitis, a possibly fatal disease. Furthermore, acute pancreatitis is four times more frequent and generally more severe in PD than in HD patients. A prospective study would be of interest, especially to investigate the specific causes of acute pancreatitis in dialysis patients.
|
Conclusions |
|---|
|
TopAbstractIntroductionPatients and methodsStatistical methodsResultsDiscussionConclusionsReferences |
|---|
Clinicians should be aware that acute abdominal pain and/or elevation of pancreatic enzymes in dialysis patients may be caused by acute pancreatitis rather than by uraemic gastritis or an insufficient renal clearance of these enzymes. Earlier diagnosis and treatment may be helpful in such patients.
Conflict of interest statement. We have nothing to declare.
|
References |
|---|
|
TopAbstractIntroductionPatients and methodsStatistical methodsResultsDiscussionConclusionsReferences |
|---|
- Baggenstoss AH. The pancreas in uremia: a histopathologic study. Am J Pathol (1948) 24:1003–1011.[Web of Science][Medline]
- Avram MM. High prevalence of pancreatic disease in chronic renal failure. Nephron (1977) 18:68–71.[Web of Science][Medline]
- Araki T, Ueda M, Ogawa K, et al. Histological pancreatitis in end-stage renal disease. Int J Pancreatol (1992) 12:263–269.[Web of Science][Medline]
- Pitrone F, Pellegrino E, Mileto G, et al. May pancreatitis represent a CAPD complication? Report of two cases with a rapidly evolution to death. Int J Artificial Organs (1985) 8:235.
- Pannekeet MM, Krediet RT, Boeschoten EW, et al. Acute pancreatitis during CAPD in the Netherlands. Nephrol Dial Transplant (1993) 8:1376–1381.
[Abstract/Free Full Text] - Rutsky EA, Robards M, Van Dyke JA, et al. Acute pancreatitis in patients with end-stage renal disease without transplantation. Arch Intern Med (1986) 146:1741–1745.
[Abstract/Free Full Text] - Caruana RJ, Wolfman NT, Karstaedt N, et al. Pancreatitis: an important cause of abdominal symptoms in patients on peritoneal dialysis. Am J Kidney Dis (1986) 7:135–140.[Web of Science][Medline]
- Padilla B, Pollak VE, Pesce A, et al. Pancreatitis in patients with end-stage renal disease. Medicine (1994) 73:8–20.[Medline]
- Van Dyke J, Rutsky EA, Stanley RJ. Acute pancreatitis associated with end-stage renal disease. Radiology (1986) 160:403–405.
[Abstract/Free Full Text] - Gupta A, Yuan ZY, Balaskas E, et al. CAPD and pancreatitis: no connection. Perit Dial Int (1992) 12:309–316.
[Abstract/Free Full Text] - Bruno MJ, van Westerloo DJ, van Dorp WT, et al. Acute pancreatitis in peritoneal dialysis and haemodialysis: risk, clinical course, outcome, and possible aetiology. Gut (2000) 46:385–389.
[Abstract/Free Full Text] - Quraishi ER, Goel S, Gupta M, et al. Acute pancreatitis in patients on chronic peritoneal dialysis: an increased risk? Am J Gastroenterol (2005) 100:2288–2293.[CrossRef][Web of Science][Medline]
- Masoero G, Bruno M, Gallo L, et al. Increased serum pancreatic enzymes in uremia: relation with treatment modality and pancreatic involvement. Pancreas (1996) 13:350–355.[Web of Science][Medline]
- Lankisch PG, Burchard-Reckert S, Lehnick D. Underestimation of acute pancreatitis: patients with only a small increase in amylase/lipase levels can also have or develop severe acute pancreatitis. Gut (1999) 44:542–544.
[Abstract/Free Full Text] - Trivedi CD, Pitchumoni CS. Drug-induced pancreatitis. An update. J Clin Gastroenterol (2005) 39:709–716.[CrossRef][Web of Science][Medline]
- Lankisch PG, Assmus C, Maisonneuve P, et al. Epidemiology of pancreatic diseases in Lüneburg county—a study in a defined German population. Pancreatology (2002) 2:469–477.[CrossRef][Web of Science][Medline]
- Asch DA, Jedrziewski MK, Christakis NA. Response rates to mail surveys published in medical journals. J Clin Epidemiol (1997) 50:1129–1136.[CrossRef][Web of Science][Medline]
Accepted in revised form: 2.10.07
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M.R. Ardalan, M.M. Shoja, K. Ghabili, S. Taheri, A. Shakeri, J. Etemadi, and B. Einollahi Fulminant Acute Pancreatitis in a Patient with Familial Mediterranean Fever on CAPD: What Caused the Pancreatitis? Perit. Dial. Int., November 1, 2009; 29(6): 676 - 678. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||