Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
REVIEW ARTICLE
Department of General, Visceral and Transplantation Surgery, University Hospital, Ulm, Germany, 2Department of
Biometry, University of Ulm, Ulm, Germany, 3Department of Surgery, Division of General Surgery, University of
Washington, Seattle, WA, USA, and 4Department of Visceral Surgery, St. Anna Virngrundklinik, Ellwangen,
Germany
Abstract
Objective. The incidence of acute pancreatitis varies from 5 to 80 per 100,000 throughout the world. The most common
cause of death in these patients is infection of pancreatic necrosis by enteric bacteria, spurring the discussion of whether or
not prophylactic antibiotic administration could be a benecial approach. In order to provide evidence of the effect of
antibiotic prophylaxis in severe acute pancreatitis (SAP) we performed an updated systematic review and meta-analysis on
this topic. Methods. The review of randomized controlled trials was performed in accordance with the Preferred Reporting
Items for Systematic Reviews and Meta-analysis (PRISMA) statement. We conducted a search of MEDLINE, EMBASE,
and the Cochrane Central Register of Controlled Trials. For assessment of the treatment effects we calculated the risk ratios
(RRs) for dichotomous data of included studies. Results. Fourteen trials were included with a total of 841 patients. The use
of antibiotic prophylaxis was not associated with a statistically signicant reduction in mortality (RR 0.74 [95% CI 0.50
1.07]), in the incidence of infected pancreatic necrosis (RR 0.78 [95% CI 0.601.02]), in the incidence of nonpancreatic infections (RR 0.70 [95% CI 0.461.06]), and in surgical interventions (RR 0.93 [95% CI 0.721.20]).
Conclusion. In summary, to date there is no evidence that supports the routine use of antibiotic prophylaxis in patients
with SAP.
Introduction
The most common cause of death in patients suffering
from severe acute pancreatitis (SAP) is the infection of
pancreatic necrosis by enteric bacteria with mortality
rates of 30% (range 1462%) [1,2], spurring the
discussion of whether or not prophylactic antibiotic
administration could be a benecial approach. Randomized controlled clinical trials (RCTs) on this topic
published in the 1990s have shown lower rates of
infected pancreatic necrosis and reduced mortality
[3,4]. Contrary to this, statistically signicant positive
effects of this prophylaxis could not be conrmed by
Correspondence: Mathias Wittau, MD, Department of General, Visceral and Transplantation Surgery, University Hospital Ulm, Steinhoevelstr. 9, 89075 Ulm,
Germany. Tel: +49 73150053500. Fax: +49 73150053503. E-mail: mathias.wittau@uniklinik-ulm.de
262
M. Wittau et al.
Methods
Data sources and study selection
A detailed protocol was developed prior to study
initiation. The review of RCTs was performed in
accordance with the Preferred Reporting Items for
Systematic Reviews and Meta-analysis (PRISMA)
statement [7]. We conducted a search of MEDLINE
(1966 to December 2009), MEDLINE in-process
and other non-indexed citations (December 2009),
EMBASE (1980 to December 2009), and the
Cochrane Central Register of Controlled Trials
(4th Quarter 2009) via PubMed and Ovid interfaces.
The following medical subject heading terms
pancreatitis, antibiotics, and randomized controlled trial and text terms were used with Boolean
operators NOT, AND and OR, as demonstrated in Appendix A for MEDLINE. Reference lists
from the trials selected by electronic searching and
conference abstracts from 2005 to 2009 (European
Pancreatic Club, American Pancreatic Association,
International Association of Pancreatology, United
European Gastroenterology Week, and Digestive Disease Week) were hand searched. Google Scholar was
used to search gray literature sources and Controlledtrials.com/ClinicalTrials.gov for ongoing registered
clinical trials. Search was performed without language
restrictions non-English/non-German manuscripts
were translated.
Inclusion criteria
.
.
.
.
Data extraction
Results
263
Records screened
(n = 1426)
Records excluded
(n = 1402)
Articles excluded:
Antibiotics vs. antibiotics (n = 5)
Antibiotic timing (n = 1)
RCT with unclear/different severity
(n = 3)
Subgroup analysis of 2 RCTs (n = 1)
Studies included in
qualitative synthesis
(n = 14)
Studies included in
quantitative synthesis
(meta-analysis)
(n = 14)
Figure 1. Flow diagram: meta-analysis of RCTs in patients with severe acute pancreatitis.
Study characteristics
Study validity
Of the 14 included RCTs (Table I), nine were singlecenter and ve mulitcenter trials. Eight trials use a
carbapenem-based prophylaxis, four studies a
1993
1995
1996
1997
2001
2002
2003
2004
2007
2007
2009
2009
2009
2009
Number
Number
Double- blinded
Number
Double- blinded
Number
Double-blinded
Number
Number
Number
Number
Number
Number
Number
Blinded
Antibiotic type
Imipenem
None
Cefuroxime
None
Ceftazidime + Amikacin + Metronidazole
None
Ooxacin + Metronidazole
None
Imipenem (primary)
Imipenem (secondary)
Ciprooxacin + Metronidazole
Ciprooxacin + Metronidazole (on demand)
Meropenem
Meropenem (on demand)
Ciprooxacin + Metronidazole
None
Imipenem
None
Meropenem
None
Imipenem
None
Ciprooxacin
None
Imipenem
None
Imipenem
None
Year
Study
41
33
30
30
11
12
13
13
25
33
33
30
20
21
58
56
36
37
50
50
24
34
22
19
29
27
28
26
n
3
4
1
7
1
3
0
2
2
5
5
3
4
5
3
4
3
4
10
9
0
0
4
2
3
4
2
3
Mortality,
n
5
10
9
12
0
3
8
7
1
6
1
0
3
6
7
5
3#
6#
9
6
3
2
8
8
8
10
6
8
Infected necrosis,
n
12
11
7
14
0
3
NR
NR
2
5
6
7
4
5
10
6
3
3
13
10
4
2
11
8
9
9
4
5
Surgery,
n
6
16
NR
NR
0
4
4
6
NR
NR
NR
NR
NR
NR
13
13
3
12
16
24
7
5
6
8
NR
NR
13
9
Non-pancreatic
infections, n
single-center
single-center
single-center
single-center
multicenter
multicenter
multicenter
multicenter
multicenter
single-center
single-center
single-center
single-center
single-center
Setting
264
M. Wittau et al.
265
Year
Adequate sequence
generation
Concealed
allocation
Blinding
Addressing incomplete
outcome data
1993
1995
1996
1997
2001
2002
2003
2004
2007
2007
2009
2009
2009
2009
Yes
Unclear
Yes
Unclear
Unclear
No
Unclear
Yes
Yes
Yes
Unclear
Yes#
Yes
Yes
Unclear
Yes
Unclear
Unclear
Unclear
Unclear
Unclear
Yes
Unclear
Yes
Unclear
Yes#
Unclear
Unclear
No
No
No
No
No
No
No
Yes
No
Yes
No
Yes
No
No
Yes
Yes
Unclear
Yes
Yes
Unclear
Unclear
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Mortality
SE (log[RR])
0.5
1.5
RR
2
0.2
Figure 2. Funnel plot: infected necrosis.
0.5
266
M. Wittau et al.
Study or Subgroup
Risk Ratio
Prophylaxis
Control
Events Total Events Total Weight M-H, Fixed, 95% Cl
Barreda et al 09
Delcenserie et al 96
Dellinger et al 07
Garcia-Barrasa 09
Isenmann et al 04
Nordback et al 01
Pederzoli et al 93
Rokke et al 07
Sainio et al 95
Schwarz et al 97
Spicak et al 02
Spicak et al 03
Xue et al 09
Yang et al 09
24
11
50
22
58
25
41
36
30
13
33
20
29
28
0
1
10
4
3
2
3
3
1
0
5
4
3
2
0
3
9
2
4
5
4
4
7
2
3
5
4
3
34
12
50
19
56
33
33
37
30
13
30
21
27
26
420
421
Total (95% Cl)
Total events
41
55
Heterogeneity: Chiz = 7.25, df = 12 (p = 0.84); Iz = 0%
Test for overall effect: Z = 1.59 (p = 0.11)
5.2%
16.2%
3.9%
7.3%
7.8%
8.0%
7.1%
12.6%
4.5%
5.7%
8.8%
7.5%
5.6%
Not estimable
0.36 [0.04, 3.00]
1.11 [0.49, 2.50]
1.73 [0.35, 8.41]
0.72 [0.17, 3.09]
0.53 [0.11, 2.50]
0.60 [0.15, 2.51]
077 [0.19, 3.20]
0.14 [0.02, 1.09]
0.20 [0.01,3.80]
1.52 [0.40, 5.81]
0.84 [0.26, 2.69]
0.70 [0.17, 2.84]
0.62 [ 0.11, 3.41]
100.0%
Risk Ratio
M-H, Fixed, 95% Cl
0.2
0.5
1
2
5
Favours Experimental Favours Control
Infected necrosis
Data of the incidence of infected necrosis were available in all trials (Figure 4). Of 841 patients, 160 developed infected pancreatic necrosis, 71/420 (17%) in
the prophylactic antibiotic group and 89/421 (21%) in
the control group. The use of antibiotic prophylaxis
Non-pancreatic infection
Nine trials reported data of the incidence of nonpancreatic infections (Figure 5) [1,3,5,6,2730,34].
A total of 563 patients were included in the analysis.
Risk Ratio
Prophylaxis
Control
Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% Cl
Barreda et al 09
Delcenserie et al 96
Dellinger et al 07
Garcia-Barrasa 09
Isenmann et al 04
Nordback et al 01
Pederzoli et al 93
Rokke et al 07
Sainio et al 95
Schwarz et al 97
Spicak et al 02
Spicak et al 03
Xue et al 09
Yang et al 09
3
0
9
8
7
1
5
3
9
8
1
3
8
6
24
11
50
22
58
25
41
36
30
13
33
20
29
28
2
3
6
8
5
6
10
6
12
7
0
6
10
8
Risk Ratio
M-H, Fixed, 95% Cl
420
421 100.0%
Total (95% Cl)
Total events
71
89
Heterogeneity: Chiz = 11.39, df = 13 (p = 0.58); Iz = 0%
Test for overall effect: Z = 1.81 (p = 0.07)
0.2
0.5
1
2
5
Favours Experimental Favours Control
Figure 4. Forest plot of relative risk; [95% condence intervals]: infected pancreatic necrosis.
Risk Ratio
Control
Risk Ratio
24
11
50
22
58
41
36
13
28
5
4
24
8
13
16
12
6
9
34
12
50
19
56
33
37
13
26
9.7%
2.0%
17.3%
11.6%
14.4%
12.2%
8.2%
9.9%
14.6%
283
280 100.0%
Total (95% Cl)
Total events
68
97
Heterogeneity: Tauz = 0.20; Chiz = 17.12, df = 8 (p = 0.03); Iz = 53%
Test for overall effect: Z = 1.67 (p = 0.10)
267
0.2
0.5
1
2
5
Favours Experimental
Favours Control
Figure 5. Forest plot of relative risk; [95% condence intervals]: non-pancreatic infections.
One hundred sixty-ve patients developed nonpancreatic infections, 68/283 (24%) in the prophylactic antibiotic group and 97/280 (35%) in the
control group. The use of antibiotic prophylaxis
was not associated with a statistically signicant
reduction in the incidence of non-pancreatic infections (random effect model, pooled RR 0.70 [95% CI
0.461.06]). There was a statistically signicant heterogeneity between the trials (p = 0.03; I2 = 53%).
Surgical intervention
Sensitivity analysis
Study or Subgroup
Risk Ratio
Prophylaxis
Control
Events Total Events Total Weight M-H, Fixed, 95% Cl
1.8%
3.7%
11.1%
9.6%
6.8%
4.8%
13.6%
3.3%
15.6%
8.2%
5.4%
10.4%
5.8%
Barreda et al 09
Delcenserie et al 96
Dellinger et al 07
Garcia-Barrasa 09
Isenmann et al 04
Nordback et al 01
Pederzoli et al 93
Rokke et al 07
Sainio et al 95
Spicak et al 02
Spicak et al 03
Xue et al 09
Yang et al 09
4
0
13
11
10
2
12
3
7
6
4
9
4
24
11
50
22
58
25
41
36
30
33
20
29
28
2
3
10
8
6
5
11
3
14
7
5
9
5
34
12
50
19
56
33
33
37
30
30
21
27
26
Figure 6. Forest plot of relative risk; [95% condence intervals]: surgical intervention.
Risk Ratio
M-H, Fixed, 95% Cl
0.2
0.5
1
2
5
Favours Experimental Favours Control
268
M. Wittau et al.
Study or Subgroup
Barreda et al 09
Delcenserie et al 96
Dellinger et al 07
Garcia-Barrasa 09
Isenmann et al 04
Nordback et al 01
Pederzoli et al 93
Rokke et al 07
Sainio et al 95
Schwarz et al 97
Spicak et al 02
Spicak et al 03
Xue et al 09
Yang et al 09
Risk Ratio
Prophylaxis
Control
Events Total Events Total Weight M-H, Fixed, 95% Cl
3
0
9
8
7
1
5
3
9
8
1
3
8
6
24
11
50
22
58
25
41
36
30
13
33
20
29
28
2
3
6
8
5
6
10
6
12
7
0
6
10
8
34
12
50
19
56
33
33
37
30
13
30
21
27
26
0.0%
0.0%
30.5%
43.6%
25.9%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
130
125
Total (95% Cl)
Total events
24
19
Heterogeneity: Chiz = 0.95, df = 2 (p = 0.62); Iz = 0%
Test for overall effect: Z = 0.63 (p = 0.53)
100.0%
Risk Ratio
M-H, Fixed, 95% Cl
0.2
0.5
1
2
5
Favours Experimental Favours Control
Figure 7. Forest plot of relative risk; [95% condence intervals]: infected pancreatic necrosis trials with low risk of bias.
Carbapenem-based prophylaxis
Eight trials used carbapenem-based prophylaxis
[3,5,27,28,31,3335]. This prophylaxis was not associated with a statistically signicant reduction in mortality (RR 0.80 [95% CI 0.501.27]), the incidence of
infected pancreatic necrosis (RR 0.69 [95% CI 0.48
1.01]), the incidence of non-pancreatic infections
(random RR 0.69 [95% CI 0.351.36]), and in surgical interventions (RR 1.00 [95% CI 0.711.40]).
269
References
[1] Isenmann R, Rnzi M, Kron M, Kahl S, Kraus D,
Jung N, et al. Prophylactic antibiotic treatment in patients
with predicted severe acute pancreatitis a placebocontrolled, double-blind trial. Gastroenterology 2004;126:
9971004.
[2] Banks PA, Freeman ML. The Practice Parameters Committee of the American College of Gastroenterology: practice
guidelines in acute pancreatitis. Am J Gastroenterol 2006;
101:2379400.
[3] Pederzoli P, Bassi C, Vesentini S, Campedelli A.
A randomized multicenter clinical trial of antibiotic prophylaxis of septic complications in acute necrotizing pancreatitis
with imipenem. Surg Gynecol Obstet 1993;176:4803.
[4] Sainio V, Kemppainen E, Puolakkainen P, Taavitsainen M,
Kivisaari L, Valtonen V, et al. Early antibiotic treatment in
acute necrotizing pancreatitis. Lancet 1995;346:6637.
[5] Dellinger EP, Tellado JM, Soto NE, Ashles SW, Barie PS,
Dugernier T, et al. Early antibiotic treatment for severe acute
necrotizing
pancreatitis:
randomized,
double-blind,
placebo-controlled study. Ann Surg 2007;245:67483.
[6] Garcia-Barrasa A, Borobia F, Pallares R, Jorba R, Poves I,
Busquets J, et al. A double-blind, placebo-controlled trial of
ciprooxacin prophylaxis in patients with acute necrotizing
pancreatitis. J Gastrointest Surg 2009;13:76874.
[7] Moher D, Liberati A, Tetzlaff J, Altman DG, The
PRISMA Group (2009). Preferred reporting items for systematic reviews and meta-analyses: The PRISMA Statement.
Open Med 2009;3(3);123130.
[8] Balthazar EJ, Robinson DL, Megibow AJ, Ranson JH. Acute
pancreatitis: value of CT in establishing prognosis. Radiology
1990;174:3316.
[9] Bradley EL III. A clinically based classication system for
acute pancreatitis. Summary of the International Symposium
on Acute Pancreatitis, Atlanta, Ga, September 11 through
13, 1992. Arch Surg 1993;128:58690.
[10] Toouli J, Brooke-Smith M, Bassi C, Carr-Locke D,
Telford J, Freeny P, et al. Working party of the program
committee of the Bangkok World Congress of Gastroenterology 2002. Guideline for the management of acute pancreatitis. J Gastroenterol Hepatol 2002;17(Suppl):s1539.
[11] Higgins JPT, Green S. Cochrane handbook for systematic
reviews of interventions 5.0.1. Chichester: JohnWiley &
Sons; 2008.
270
M. Wittau et al.
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
Copyright of Scandinavian Journal of Gastroenterology is the property of Taylor & Francis Ltd and its content
may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express
written permission. However, users may print, download, or email articles for individual use.