| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
Medicina d'Urgenza e Pronto Soccorso, Dipartimento di Medicina Interna e Gastroenterologia, Laboratorio Centralizzato, Sant' Orsola Hospital, 40138 Bologna, Italy.
a Address correspondence to this author at: Medicina d'Urgenza e Pronto Soccorso, Ospedale Sant'Orsola, 40138 Bologna, Italy. Fax 39-051-6364794; e-mail pezzilli{at}altavista.net
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Methods: Eighty patients with acute abdomen were studied on admission to the Emergency Room. Forty patients had nonpancreatic acute abdomen, and 40 had acute pancreatitis (25 had mild acute pancreatitis and 15 had severe pancreatitis). Forty healthy subjects comparable for sex and age were also studied as controls. Lipase, interleukin-6, and C-reactive protein were determined on serum in all subjects.
Results: Using lipase to discriminate between patients with nonpancreatic acute abdomen and patients with acute pancreatitis (cutoff values ranging from 419 to 520 U/L), one patient with acute pancreatitis was not identified correctly. To discriminate between patients with severe acute pancreatitis and those with mild pancreatitis in the remaining 39 patients, interleukin-6 (cutoff value, <3.7 µg/L) had a sensitivity of 100% (15 of 15) and a specificity of 83% (20 of 24); 75 of 80 (94%) patients were classified correctly. C-reactive protein (cutoff values ranging from 6 to 7 mg/L) showed a lower prognostic efficiency than interleukin-6: sensitivity of 87% (13 of 15) and specificity of 46% (11 of 24). Sixty-four of 80 patients (80%) were classified correctly. The area under the ROC curve for interleukin-6 (0.911 ± 0.049) was significantly (P = 0.013) greater than that for C-reactive protein (0.685 ± 0.090).
Conclusion: The combined use of serum lipase and interleukin-6 is useful in simultaneously establishing both the diagnosis and the prognosis of acute pancreatitis.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
.
|
Forty patients admitted to the Emergency Room in the same period
(August 1997 through April 1998) with a diagnosis of acute abdomen of
nonpancreatic origin were also studied (20 men and 20 women; mean age,
53 years; range, 22–97 years). For each acute pancreatitis patient
admitted to the study, one patient with an acute abdomen of
nonpancreatic origin was enrolled. Final diagnoses are listed in Table 2
.
|
Patients who had not given informed consent, those with psychoses, as well as those with a history of malignancies, chronic renal and hepatic insufficiency, and chronic pancreatitis were excluded from the study. Forty healthy subjects, randomly selected in the same time period from blood donors and subjects who underwent medical check-ups, were also studied; there were 26 men and 14 women (mean age, 54 years; range, 18–82 years). They had been screened by means of an interview, physical examination, and routine blood analysis and showed no evidence of illness. No significant differences in gender and age were found among the three groups of subjects studied.
After all subjects had given informed consent, a blood sample was obtained from each patient in the Emergency Room at the initial visit and from each healthy subject the morning after an overnight fast. The sera were kept frozen at -20 °C until analysis.
Lipase, IL-6, and CRP were determined on the serum of all subjects studied using commercially available reagents. Assays were performed without knowledge of the patients' findings or other test results. Lipase was determined using a turbidimetric assay (DuPont; reference interval, 24–270 U/L) (9). The lower limit of detection was 5 U/L, and the calibration curve ranged from 0 to 1200 U/L; the mean recovery was 102%. The within-run imprecision (CV, n = 20) was 3.3% at 270 U/L and 2.8% at 810 U/L, whereas the day-to-day imprecision (10 determinations) was 7.3% at 270 U/L and 6.5% at 810 U/L. Serum IL-6 was determined using an ELISA kit (Predicta IL-6 ELISA kit; Genzyme Co.). The lower limit of detection was 2 µg/L, and the calibration curve ranged from 0 to 512 µg/L; the mean recovery was 102%. The within-run imprecision (CV, n = 20) was 8% at 60 µg/L and 4% at 300 µg/L, whereas the day-to-day imprecision (10 determinations) was 10% at 60 µg/L and 5.4% at 810 µg/L. With this assay, healthy subjects had no detectable IL-6 in serum (4). CRP was determined using a nephelometric technique (Beckman, Inc.; reference interval, 0–8 mg/L) (10). The lower limit of detection was 1 mg/L, and the calibration curve ranged from 1 to 120 mg/L; the mean recovery of added CRP was 103%. The within-run imprecision (CV, n = 20) was 3% at 10 mg/L and 2% at 60 mg/L, whereas the day-to-day imprecision (10 determinations) was 7% at 10 mg/L and 4% at 60 mg/L.
statistical analysis
ROC curves and the respective areas under the curves were
calculated for each marker to provide information about the accuracy of
serum lipase in distinguishing between patients with acute pancreatitis
and those with nonpancreatic acute abdomen and about the accuracy of
both serum IL-6 and CRP in distinguishing severe acute pancreatitis
from the mild form of the disease. Nonparametric estimates of the areas
under the ROC curves and the respective SEs were applied
(11). A procedure involving paired data was applied to
compare areas under the ROC curves (12). The best cutoff was
chosen as the value that maximized the likelihood ratio (LR) obtained
using the following formula: LR = (probability of true positive +
probability of true negative)/ (probability of false positive +
probability of false negative) (4). The best cutoff values
are reported as ranges in which the diagnostic performance of the
method does not change. Using these cutoff values, we calculated the
sensitivity, specificity, and the percentage of correct results in
simultaneously establishing the diagnosis (acute pancreatitis patients
vs patients with nonpancreatic acute abdomen) and prognosis (severe vs
mild acute pancreatitis patients) of acute pancreatitis for serum
lipase, IL-6, and CRP. The exact 95% confidence intervals (CIs) of
these proportions were also evaluated (13). The
Yates-corrected 
2 and the Mann-Whitney
U-test were applied to compare different groups of subjects.
All statistical evaluations were performed running the SPSS/PC+
statistical package on a personal computer (14). A
two-tailed P <0.05 was used to define statistical
significance.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
. Serum lipase concentrations were higher than the upper limit
of normal in all patients with acute pancreatitis, in three patients
with acute abdomen of other etiologies (7.5%; CI, 1.6–20%; one
patient had a duodenal ulcer, one had gallstones, and one had an
intestinal obstruction because of an inguinal hernia), and in two
healthy subjects (5%; CI, 0.6–17%). Serum IL-6 concentrations were
abnormally high in all (100%; CI, 78–100%) of the 15 patients with
severe acute pancreatitis, in 4 patients with mild acute pancreatitis
(16%; CI, 4.5–36%), and in 30 patients with nonpancreatic acute
abdomen (75%; CI, 59–87%). Serum CRP concentrations were abnormally
high in 12 patients with severe acute pancreatitis (80%; CI,
52–96%), in 13 patients with mild acute pancreatitis (52%; CI,
31–72%), and in 15 patients with nonpancreatic acute abdomen (38%;
CI, 23–54%).
|
As shown in Table 3
, patients with severe acute pancreatitis had significantly
higher serum concentrations of IL-6 than patients with mild acute
pancreatitis (P <0.001) and healthy subjects (P
<0.001), whereas no significant differences were found in comparison
with patients having nonpancreatic acute abdomen. Patients with mild
acute pancreatitis had significantly higher serum concentrations of
IL-6 (P = 0.009) than those of healthy subjects but
significantly lower serum concentrations (P <0.001) than
those of nonpancreatic acute abdomen. Patients with nonpancreatic acute
abdomen had serum concentrations of this protein significantly higher
(P <0.001) than those of healthy subjects.
|
Regarding serum CRP, patients with severe acute pancreatitis had serum concentrations of this protein that were not significantly different from patients with mild acute pancreatitis but significantly higher than patients with nonpancreatic acute abdomen (P = 0.006) and healthy subjects (P <0.001). Both the patients with mild acute pancreatitis and those with nonpancreatic acute abdomen had significantly higher serum concentrations of CRP (P <0.001 and P = 0.026, respectively) than those of healthy subjects; patients with nonpancreatic acute abdomen also had serum concentrations of CRP significantly higher (P = 0.049) than those of patients with mild acute pancreatitis.
The ROC curves for serum lipase in differentiating between patients
with acute pancreatitis and those with nonpancreatic acute abdomen, and
for serum IL-6 and CRP in differentiating between the severe form and
the mild form of acute pancreatitis, are shown in Fig. 2
. The values of the areas under the ROC curves (±SE) are
0.999 ± 0.002 for serum lipase, 0.911 ± 0.049 for serum
IL-6, and 0.658 ± 0.090 for serum CRP. The area of serum CRP was
significantly lower (P = 0.013) than that of serum
IL-6. The best cutoff value of serum lipase to differentiate between
patients with acute pancreatitis and those patients with nonpancreatic
acute abdomen ranged from 419 to 520 U/L. The best cutoff value to
differentiate severe acute pancreatitis from the mild form of the
disease was <3.7 µg/L for IL-6 and ranged from 6 to 7 mg/L for CRP.
In using lipase to discriminate between patients with nonpancreatic
acute abdomen and patients with acute pancreatitis, one patient with
mild acute pancreatitis, who had serum concentrations of lipase equal
to 337 U/L, of IL-6 <2 µg/L, and of CRP equal to 4 mg/L, was not
identified correctly. In the remaining 39 patients, to distinguish
those with severe acute pancreatitis from those with the mild form of
the disease, a cutoff value <3.7 µg/L for IL-6 had a sensitivity of
100% (15 of 15 severe acute pancreatitis; CI, 78–100%) and a
specificity of 83% (20 of 24 mild acute pancreatitis; CI, 63–95%). A
total of 94% of the cases (severe acute pancreatitis + mild acute
pancreatitis + nonpancreatic acute abdomen: 75 of 80; CI, 86–98%)
were classified correctly. CRP (cutoff between 6 and 7 mg/L) showed a
lower prognostic efficiency than IL-6: sensitivity of 87% (13 of 15;
CI, 60–99%) and specificity of 46% (11 of 24; CI, 26–67%); 80% of
the cases (severe acute pancreatitis + mild acute pancreatitis +
nonpancreatic acute abdomen: 64 of 80; CI, 70–88%) were classified
correctly.
|
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
At present, many systems are used in determining the prognosis of acute pancreatitis (1)(2)(4)(17); these systems require clinical, laboratory, and radiological assessment of the disease, and none of them are useful for a rapid evaluation of high risk patients in emergency situations because they are expensive or time-consuming. It has been reported that serum concentrations of IL-6 increase in patients with inflammatory disorders (18)(19)(20)(21)(22)(23)(24)(25) and in patients with acute pancreatitis (4)(26)(27)(28)(29); however, there are no studies that used this molecule to establish the prognosis of acute pancreatitis at the first initial visit when the patient was in the Emergency Room. We have also found recently that the combined use of serum lipase either with CRP or with ß2-microglobulin is not useful in simultaneously establishing the diagnosis and the prognosis of acute pancreatitis (3). In the present study, we evaluated whether the combined use of serum lipase and serum IL-6 assays, i.e., two tests that are quick and easy to perform, is able to serve this purpose. We also compared the sensitivity and the specificity of this combination with that of serum lipase and serum CRP.
Patients with severe acute pancreatitis, those with the mild form of the disease, and those with nonpancreatic acute abdomen had serum concentrations of both IL-6 and CRP significantly higher than those of the healthy subjects. These findings confirm previous studies (4)(5)(19)(21)(26) showing that in patients with acute pancreatitis and in those with nonpancreatic acute abdomen, there is early involvement of the immune system.
The combined use of serum lipase and IL-6 allows us to simultaneously establish the diagnosis and the prognosis of acute pancreatitis in 94% of patients with acute abdomen, whereas the combined use of serum lipase and CRP had a lower performance (80%). These results may be because of the fact that, in acute pancreatitis patients, there is an activation of the acute phase response that is more pronounced in the severe form of the disease and involves proinflammatory cytokines such as IL-6 at an early stage (27). On the contrary, CRP only indicates the severity of acute pancreatitis at a later stage because it requires the activation of cytokines for its synthesis by the liver (28)(29).
In conclusion, the results of this study show that in emergency situations, it is now possible to simultaneously establish both the diagnosis and the prognosis of acute pancreatitis to put the high-risk patients in the intensive care unit and to immediately initiate intensive medical treatment. We hope that an IL-6 assay for routine use in emergency situations will be developed in the near future. Additional clinical studies involving a more consistent population of patients with acute pancreatitis are needed to confirm these initial and promising results.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
, interleukin-6, parameters of the cellular immune system in patients with intraabdominal sepsis. Scand J Infect Dis 1992;24:361-368.
[ISI][Medline]
[Order article via Infotrieve]
and interleukin 6 plasma levels in infected cirrhotic patients. Gastroenterology 1993;104:1492-1497.
[ISI][Medline]
[Order article via Infotrieve]
in ascitic fluid during spontaneous bacterial peritonitis. Dig Dis Sci 1991;36:123-124.
[ISI][Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  W R Matull, S P Pereira, and J W O'Donohue Biochemical markers of acute pancreatitis. J. Clin. Pathol., April 1, 2006; 59(4): 340 - 344. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A Fabbri, F Servadei, G Marchesini, A M Morselli-Labate, M Dente, T Iervese, M Spada, and A Vandelli Prospective validation of a proposal for diagnosis and management of patients attending the emergency department for mild head injury J. Neurol. Neurosurg. Psychiatry, March 1, 2004; 75(3): 410 - 416. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Shimada, A. Andoh, K. Hata, K. Tasaki, Y. Araki, Y. Fujiyama, and T. Bamba IL-6 Secretion by Human Pancreatic Periacinar Myofibroblasts in Response to Inflammatory Mediators J. Immunol., January 15, 2002; 168(2): 861 - 868. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-L. FROSSARD, A. HADENGUE, and C. M. PASTOR New Serum Markers for the Detection of Severe Acute Pancreatitis in Humans Am. J. Respir. Crit. Care Med., July 1, 2001; 164(1): 162 - 170. [Full Text] [PDF]
|
|
|
|
 |
|
 J Mayer, B Rau, F Gansauge, and H G Beger Inflammatory mediators in human acute pancreatitis: clinical and pathophysiological implications Gut, October 1, 2000; 47(4): 546 - 552. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
