Original article

Multicentre trial of a perioperative protocol to reduce mortality

in patients with peptic ulcer perforation
M. H. Møller1 , S. Adamsen2 , R. W. Thomsen3 and A. M. Møller1 on behalf of the Peptic Ulcer
Perforation (PULP) trial group
Departments of 1 Anaesthesiology and Intensive Care Medicine and 2 Gastrointestinal Surgery, Copenhagen University Hospital Herlev, Herlev, and
3
Department of Clinical Epidemiology, Clinical Institute, Aarhus University Hospital, Aarhus, Denmark
Correspondence to: Dr M. H. Møller, Department of Anaesthesiology and Intensive Care Medicine, Copenhagen University Hospital Herlev, Herlev
Ringvej 75, DK – 2730 Herlev, Denmark (e-mail: mortenhylander@gmail.com)

Background: Morbidity and mortality rates in patients with perforated peptic ulcer (PPU) remain
substantial. The aim of the present study was to evaluate the effect of a multimodal and multidisciplinary
perioperative care protocol on mortality in patients with PPU.
Methods: This was an externally controlled multicentre trial set in seven gastrointestinal departments
in Denmark. Consecutive patients who underwent surgery for gastric or duodenal PPU between
1 January 2008 and 31 December 2009 were treated according to a multimodal and multidisciplinary
evidence-based perioperative care protocol. The 30-day mortality rate in this group was compared with
rates in historical and concurrent national controls.
Results: The 30-day mortality rate following PPU was 17·1 per cent in the intervention group, compared
with 27·0 per cent in the three control groups (P = 0·005). This corresponded to a relative risk of 0·63
(95 per cent confidence interval 0·41 to 0·97), a relative risk reduction of 37 (5 to 58) per cent and a
number needed to treat of 10 (6 to 38).
Conclusion: The 30-day mortality rate in patients with PPU was reduced by more than one-third after
the implementation of a multimodal and multidisciplinary perioperative care protocol, compared with
conventional treatment. Registration number: NCT00624169 (http://www.clinicaltrials.gov).

Paper accepted 16 December 2010

Published online 25 March 2011 in Wiley Online Library (www.bjs.co.uk). DOI: 10.1002/bjs.7429

Introduction To improve the outcome of patients with PPU, an

Perforated peptic ulcer (PPU) is a complication of peptic
ulcer disease in which gas and gastroduodenal fluid leak
into the peritoneal cavity. The lifetime prevalence of PPU
has been estimated at 5 per cent1 . Mortality and morbid-
ity following PPU are substantial, and mortality rates as
high as 25–30 per cent have been reported2 – 5 . Several
prognostic factors and scoring systems for PPU have been
examined4,6 – 10 . Sepsis is frequent and a leading cause of
death in patients with PPU, and some of the reported prog-
nostic factors are associated with the sepsis syndrome11,12 .
Recently, fast-track surgery and multimodal periopera-
tive care protocols have gained popularity13 – 15 . Through
the use of evidence-based, often multidisciplinary, inter-
ventions – both before, during and after surgery – quality
of care has been improved, with documented reductions
in morbidity and mortality rates among selected surgical
patients14 .
evidence-based perioperative care protocol is needed. The
aim of the present study was to evaluate the effect of
a multimodal and multidisciplinary perioperative care
protocol in patients with PPU in seven gastrointestinal
departments in Denmark. It was hypothesized that a
perioperative care protocol, based on evidence-based
interventions and the Surviving Sepsis Campaign16 , could
reduce postoperative morbidity and mortality associated
with PPU.
Methods
This controlled multicentre intervention clinical trial
was conducted using historical and concurrent national
controls. The trial was approved by the Danish National
Committee on Biomedical Research Ethics; because the
intervention consisted of currently used and evidence-
based treatments, the trial was approved as a quality audit

 2011 British Journal of Surgery Society Ltd British Journal of Surgery 2011; 98: 802–810
Published by John Wiley & Sons Ltd
Optimized perioperative care in perforated peptic ulcer 803

without obligation to notify. The trial was registered at Table 1 Definition of the intervention group and the three
clinicaltrials.gov (NCT00624169). control groups

All hospitals in
Participants and settings Seven hospitals Denmark, excluding
participating in the seven PULP
A total of 206 consecutive patients admitted to seven PULP trial trial hospitals

gastrointestinal departments in Denmark with suspected Pre-PULP trial period Historical control Historical national
PPU between 1 January 2008 and 31 December 2009 2003–2007 group control group
PULP trial period Intervention group Concurrent national
were assessed for eligibility. The participating hos-
2008–2009 control group
pitals were: Copenhagen University Hospital Herlev,
Copenhagen University Hospital Bispebjerg, Copen- PULP, Peptic Ulcer Perforation.
hagen University Hospital Hillerød, Copenhagen Uni-
versity Hospital Helsingør, Aarhus University Hospital,
Odense University Hospital and Kolding Hospital. Preg- Assessed for eligibility n = 206
nant and breastfeeding women, patients who received
conservative treatment, those with malignant ulcers and
patients with perforation of other organs were excluded. Excluded n = 33
Non-surgical treatment n = 2
Patients for whom PPU could not be confirmed dur- Not enrolled by participating centres
ing surgery (for example patients with perforation of for various reasons n = 31
organs other than the stomach or duodenum) were
excluded; the intervention protocol was discontinued in
these patients. Received allocated intervention n = 173

Data extraction and management Discontinued intervention n = 56

During operation n = 55
Primary data extraction was performed by principal Colonic perforation n = 29
investigators at each centre using a standard case report Small bowel perforation n = 9
form. Data were subsequently validated and transferred Necrotizing pancreatitis n = 3
Perforated appendix n = 6
to an electronic database by the principal author. Other perforation n = 8
The manuscript has been prepared according to the After surgery n = 1
CONSORT statement17,18 . Perforation due to gastric
malignancy n = 1

Trial protocol Analysed n = 117

A perioperative care protocol was implemented for all

patients with suspected PPU at the intervention centres.
The aim was to improve core areas in the treatment of
PPU throughout the entire hospital stay: before, during
and after surgery. Surgeons, anaesthesiologists, emergency
room nurses, anaesthesiology nurses, scrub nurses, inten-
sive care nurses and nurses in the recovery room were
instructed in use of the protocol and participated in
the trial. The intervention consisted of basic and well
established diagnostic and treatment principles: evalu-
ation and risk stratification by experienced healthcare
professionals; minimization of surgical delay; early use of
broad-spectrum empirical antibiotics; standard blood tests
and electrocardiography; respiratory and circulatory stabi-
lization in a high-dependency unit; relevant antisecretory
therapy; focus on administration of nutrition and fluids
after surgery; appropriate analgesia; early mobilization;
Fig. 1CONSORT diagram showing patient enrolment and
follow-up
prevention of atelectasis and other postoperative compli-
cations; and frequent and sufficient monitoring of vital
parameters. The individual elements of the trial proto-
col were decided by consensus agreement in the author
group.
Outcome measures
The primary outcome measure was 30-day mortality.
Secondary outcome measures were reoperation (a surgical
procedure related to the primary surgical procedure) within
30 days of surgery, and total length of hospital stay.
In the intervention group, the following three tertiary

 2011 British Journal of Surgery Society Ltd www.bjs.co.uk British Journal of Surgery 2011; 98: 802–810
Published by John Wiley & Sons Ltd
804 M. H. Møller, S. Adamsen, R. W. Thomsen and A. M. Møller

Table 2 Compliance with the trial protocol

No. of patients
(n = 117)

Before surgery
Evaluation by a consultant anaesthetist and a consultant surgeon with discussion of therapeutic options 104 (88·9)
Surgery within 6 h of admission 74 (63·2)
APACHE II scoring33 * 97 (82·9)
Sepsis screening20 105 (89·7)
Standard blood samples and electrocardiogram 110 (94·0)
Packed red blood cells if patient anaemic 23 (19·7)
Broad-spectrum empirical antibiotics 113 (96·6)
Measurement of bodyweight 108 (92·3)
Respiratory and circulatory stabilization in the high-dependency unit 105 (89·7)
Oxygen treatment leading to arterial oxygen saturation ≥ 94%
Insertion of central venous catheter, arterial line, NGT and indwelling bladder catheter
Early goal-directed fluid therapy34
During surgery
Fluid balance chart 106 (90·6)
Early goal-directed fluid therapy34 99 (84·6)
Maintenance of normothermia using convective air warming system 112 (95·7)
Insertion of double-barrelled NGT 89 (76·1)
After surgery
Evaluation by a consultant anaesthetist and a consultant surgeon; postop. treatment planned 94 (80·3)
Early goal-directed fluid therapy34 in recovery room 106 (90·6)
Oxygen treatment leading to arterial oxygen saturation ≥ 94% on postop. days 1–2 111 (94·9)
Sepsis screening20 daily on postop. days 1–3 83 (70·9)
Standard systemic analgesia with opioids + paracetamol (acetaminophen), + continuous epidural analgesia, if indicated 117 (100)
Chest physiotherapy on postop. days 1–3 99 (84·6)
Standard blood samples and electrocardiogram on postop. days 1–3 105 (89·7)
Proton pump inhibitor treatment 115 (98·3)
Evaluation by anaesthesiologist if postop. condition deteriorates in recovery room 117 (100)
Postop. plan of nutrition within 12 h of admission 103 (88·0)
≥ 12 h in recovery room 110 (94·0)
Removal of central venous catheter, arterial line and indwelling bladder catheter when no longer indicated 117 (100)
Written plan of treatment and monitoring in first 24 h in regular surgical ward after discharge from recovery room 97 (82·9)
Monitoring of blood pressure, heart rate, respiratory rate, oxygen saturation and level of consciousness 3 times daily on postop. 63 (53·8)
days 1–3
Fluid balance chart on postop. days 1–3 101 (86·3)
Bodyweight measurement daily on postop. days 1–3 62 (53·0)
Early mobilization 115 (98·3)
Use of medical emergency team-calling criteria in regular surgical department35 12 (10·4)

Values in parentheses are percentages. *Mean(s.d.) Acute Physiology And Chronic Health Evaluation (APACHE) II score 14·0(6·4). NGT, nasogastric
tube.

outcomes were also registered: development of pneumonia
(according to Centers for Disease Control and Prevention
criteria19 ) within 30 days of surgery; development of
septic shock (according to the 2001 International Sepsis
Conference20 ) within 30 days of surgery; and postoperative
admission to the intensive care unit (ICU) within 30 days
of surgery.
Control groups
The intervention group (the Peptic Ulcer Perforation
(PULP) trial cohort) was compared with three control
cohorts21 : a concurrent national control cohort of patients
with PPU admitted to any hospital in Denmark except
the seven PULP trial hospitals during 2008–2009; a
historical control cohort of patients with PPU admitted
to the same seven PULP trial hospitals during 2003–2007;
and a historical national control cohort of patients with
PPU admitted to any hospital in Denmark except the
seven PULP trial hospitals during 2003–2007 (Table 1).
Inclusion and exclusion criteria in the three control
groups were identical to those in the intervention group.
Control group data were retrieved from the Danish
National Indicator Project database (http://www.nip.dk),
which holds prospectively collected information on all
patients surgically treated for confirmed PPU in Denmark.
The Danish National Indicator Project database is a

 2011 British Journal of Surgery Society Ltd www.bjs.co.uk British Journal of Surgery 2011; 98: 802–810
Published by John Wiley & Sons Ltd
Optimized perioperative care in perforated peptic ulcer 805

Table 3 Demographic characteristics of the intervention group compared with the three control groups

Intervention Concurrent national Historical Historical national

group control group control group control group
(n = 117)* (n = 512)‡ (n = 510)§ (n = 1480)¶

n* P†† n* P†† n* P††

Age (years)† 70 (25–92) 71 (18–101) 0·850‡‡ 69 (16–98) 0·850‡‡ 71 (17–104) 0·850‡‡

Men 57 (48·7) 239 (46·7) 0·708 239 (46·9) 0·708 643 (43·4) 0·261
Body mass index (kg/m2 )† 24 (15–65) 23 (14–65) 0·819‡‡ 23 (15–48) 0·819‡‡ 23 (12–57) 0·819‡‡
Co-morbidity# 85 (72·6) 353 (68·9) 0·445 334 (65·5) 0·182 988 (66·8) 0·252
ASA grade
I 12 (10·3) 92 (18·0) 0·008 107 (21·0) < 0·001 326 (22·0) < 0·001
II 43 (36·8) 190 (37·1) 1·000 163 (32·0) 0·320 488 (33·0) 0·427
III 45 (38·5) 169 (33·0) 0·324 158 (31·0) 0·167 488 (33·0) 0·324
IV 15 (12·8) 56 (10·9) 0·556 77 (15·1) 0·556 163 (11·0) 0·556
V 2 (1·7) 5 (1·0) 0·476 5 (1·0) 0·476 15 (1·0) 0·476
Daily smoking 59 (50·4) 297 (58·0) 0·135 332 (65·1) 0·005 932 (63·0) 0·015
Alcohol abuse** 30 (25·6) 92 (18·0) 0·075 122 (23·9) 0·657 266 (18·0) 0·075

*Values in parentheses are percentages unless indicated otherwise; †values are median (range). ‡Denmark exclusive of the Peptic Ulcer Perforation
(PULP) trial hospitals, 2008–2009; §the seven hospitals in the intervention group (PULP trial hospitals), 2003–2007; ¶Denmark exclusive of the PULP
trial hospitals, 2003–2007. #One or more of the following: diabetes, chronic obstructive pulmonary disease, heart disease, liver cirrhosis, active cancer
disease, acquired immune deficiency syndrome or other chronically treated disease. **More than 36 g/day (men) or 24 g/day (women). ASA, American
Society of Anesthesiologists. ††Versus intervention group (χ2 test, except ‡‡independent-samples t test).

Table 4 Perioperative data for the intervention group and the three control groups

Intervention Concurrent Historical Historical national

group national control control group control group
(n = 117) group (n = 512)† (n = 510)‡ (n = 1480)§

Time to admission (h)* 10 (1–336) — — —

Time from admission to surgery (h)* 5 (1–72) — — —
Duration of surgery (min)* 83 (30–240) — — —
Surgical procedure — — —
Simple open closure ± omental patch 97 (82·9)
Laparoscopic closure 12 (10·3)
Resection 8 (6·8)
Gastric site of perforation¶ 42 (35·9) 266 (52·0) 275 (53·9) 784 (53·0)
Preoperative metabolic acidosis 56 (47·9) — — —
Preoperative haemoglobin < 6.0 mmol/l 10 (8·5) 56 (10·9) 61 (12·0) 178 (12·0)
Preoperative creatinine > 110/130 µmol/l (W/M) 33 (28·2) 133 (26·1) 148 (29·0) 385 (26·0)
Preoperative albumin < 550 µmol/l# 56 (47·9) — 362 (71·0) 1061 (71·7)
Use of NSAIDs at admission 44 (37·6) 169 (33·0) 230 (45·1) 651 (44·0)
Use of aspirin at admission 36 (30·8) 138 (27·0) 138 (27·1) 459 (31·0)
Use of steroids at admission 17 (14·5) 51 (10·0) 66 (12·9) 222 (15·0)
Use of anticoagulants at admission 7 (6·0) 31 (6·1) 26 (5·1) 74 (5·0)
Shock on admission** 28 (23·9) 118 (23·0) 71 (13·9) 192 (13·0)

Values in parentheses are percentages unless indicated otherwise; *values are median (range). †Denmark exclusive of the Peptic Ulcer Perforation
(PULP) trial hospitals, 2008–2009; ‡the seven hospitals in the intervention group (PULP trial hospitals), 2003–2007; §Denmark exclusive of the PULP
trial hospitals, 2003–2007. ¶In the control groups, site of perforation was divided into three categories (gastric, other and unknown), whereas in the
intervention group there were only two categories (gastric and other). #Only registered until 2007. **Way of registering shock changed in 2007. —, Not
registered in control groups; W/M, women/men; NSAID, non-steroidal anti-inflammatory drug.

nationwide initiative to document, monitor and improve hospital departments caring for patients with PPU
the quality of treatment and care provided by the participated in the Danish National Indicator Project
Danish healthcare system22 . Reporting of patients with during the study interval. The proportion of patients dying
surgically verified PPU is mandatory, and all Danish within 30 days following surgery is a standard element

 2011 British Journal of Surgery Society Ltd www.bjs.co.uk British Journal of Surgery 2011; 98: 802–810
Published by John Wiley & Sons Ltd
806
and quality of care indicator in the database, and is
established by linkage with the Danish Civil Registration
System.
Statistical analysis
The expected baseline 30-day mortality rate following
surgery for PPU was 31 per cent, based on previous obser-
vations in Denmark23 . Based on international literature,
with a similar patient case mix as the Danish2,3,24 – 30 , the
expected 30-day mortality rate in the intervention group
was 18 per cent. This corresponded to a relative risk reduc-
tion of 42 per cent. Based on 80 per cent power to detect
a possible significant difference (P = 0·050, two-sided)
between two groups with a 4 : 1 ratio of control patients
versus those receiving the intervention, 110 patients were
required in the intervention group and 400 in the con-
trol group.
The intervention group was compared with the three
control groups using the independent-samples t test and
χ2 test. Results are presented as proportions, relative
risks, relative risk reduction and number needed to treat
(NNT) with 95 per cent confidence intervals. Two-sided
P < 0·050 was considered statistically significant. Data
were analysed using SPSS version 18.0 (SPSS, Chicago,
Illinois, USA) statistical software package.
Data were missing for less than 5 per cent of the
patients. The prevalence and pattern of missing values
were evaluated, and data were found not to be missing
at random. Consequently, multiple imputation for the
missing values was performed31,32 . Five imputations were
made, and the five data sets were analysed by using Markov
chain Monte Carlo fully conditional specification31 ,
and the estimates were pooled for overall estimation.
Furthermore, complete-case analysis was performed. The
pooled imputed estimates and the corresponding complete-
case estimates were compared. If there were any noticeable
differences, both results are presented; otherwise, only the
pooled imputed estimates are presented.
Results
A total of 117 patients who had surgical treatment for
gastric or duodenal PPU were included in the intervention
group. A further 31 patients who were treated surgically for
PPU in the intervention period were not enrolled primarily
owing to logistical difficulties and lack of resources (Fig. 1).
Characteristics of the trial protocol and compliance with
each element are shown in Table 2.
Demographic characteristics of the intervention and
control groups are presented in Table 3. Patients in the
 2011 British Journal of Surgery Society Ltd
Published by John Wiley & Sons Ltd
M. H. Møller, S. Adamsen, R. W. Thomsen and A. M. Møller
Intervention group, 2008–2009
Concurrent national control group, 2008–2009
Historical control group, 2003–2007
35 Historical national control group, 2003–2007
30
25
30-day mortality (%)
20
15
10
5
0
Fig. 2Thirty-day mortality rate with 95 per cent confidence
interval in the intervention group (117 patients) compared with
the concurrent national control group (512 patients), the
historical control group in the intervention centres (510 patients)
and the historical national control group (1480 patients).
*P = 0·005 versus control groups (χ test)
2
intervention group were similar to those in the three
control groups in terms of age distribution, coexisting
disease and body mass index. A higher proportion of
patients were classified as previously healthy (American
Society of Anesthesiologists (ASA) grade I) in the control
groups than in the intervention group. There were more
smokers in the two historical control groups. Perioperative
characteristics of the intervention group and the three
control groups are shown in Table 4.
The 30-day mortality rate was 17·1 per cent (20 of 117)
in the intervention group, compared with 27·0 per cent in
the control groups (138 of 512, 138 of 510 and 400 of 1480;
P = 0·005) (Fig. 2). This corresponded to a relative risk of
0·63 (0·42 to 0·97) compared with concurrent national con-
trols, 0·63 (0·41 to 0·97) compared with historical controls
in the same centres, and 0·63 (0·42 to 0·95) compared with
historical national controls. The relative risk reduction was
37 (5 to 58) per cent, and the NNT to avoid one death was
10 (6 to 36). Annual 30-day mortality following PPU from
2003 to 2009 is shown in Fig. 3.
Among the 31 patients with PPU who were eligible
for treatment but did not receive the intervention (those
www.bjs.co.uk British Journal of Surgery 2011; 98: 802–810
Optimized perioperative care in perforated peptic ulcer
All Danish hospitals
Danish hospitals except PULP trial hospitals
35 PULP trial hospitals
30
25
30-day mortality (%)
20
15
10
5 PULP centres during 2008–2009 was independent of the
0
2003 2004 2005 2006 2007 2008
Fig. 3Thirty-day mortality from 2003 to 2009 in all Danish
hospitals, all Danish hospitals except Peptic Ulcer Perforation
(PULP) trial hospitals, and PULP trial hospitals. *The 117
patients in the intervention group
missed for enrolment by the PULP centres), the 30-day
mortality rate was 23 per cent. These patients were similar
to those in the intervention group with respect to age,
coexisting disease, ASA grade and sex.
Some 17·1 per cent of patients in the intervention
group (20 of 117) required reoperation, compared with
15·0 per cent (77 of 512) in the concurrent national
control group (P = 0·530), 19·0 per cent (97 of 510) in
the historical PULP centre control group (P = 0·620),
and 19·0 per cent (281 of 1480) in the historical national
control group (P = 0·620). Median length of hospital stay
in the intervention group was 10 (interquartile range 6–20)
days, compared with 8 days in all three control groups
(P = 0·030).
In the intervention group, 27 patients (23·1 per cent)
developed pneumonia19 , 30 (25·6 per cent) were diagnosed
with septic shock20 and 49 (41·9 per cent) were admitted
to the ICU.
Discussion
In this controlled multicentre clinical trial with historical
and concurrent national control groups, the implementa-
tion of a multimodal and multidisciplinary perioperative
care protocol in patients with PPU was associated with
a substantial and statistically significant reduction in the
30-day mortality rate.
This trial has several limitations. A randomized trial
would have been the standard of excellence for determining
the effect of the intervention, but blinding would have
been difficult. Randomization was not carried out for
 2011 British Journal of Surgery Society Ltd
Published by John Wiley & Sons Ltd
807
ethical reasons and because of a high risk of an unintended
intervention effect in the control group (spillover effect).
Trials with historical controls often find the intervention
to be better because of either general improvements in
therapy, or diagnosis of less severe disease over time36 – 38 .
However, the interval between the selection of historical
controls and the intervention group in this study was
relatively short, and no improvement in PPU prognosis
was noted either at national level or in the intervention
centres between 2003 and 2007 (Fig. 3). This makes it
unlikely that the significant reduction in mortality in the
intervention.
2009 Both time and place are usually related to prognosis,
and referral patterns and skill of staff may vary between
hospitals37 . To mimimize bias, the study included both
population-based concurrent controls treated during the
same time interval in the rest of Denmark and historical
controls from the same setting21 , with identical results.
Moreover, the Danish National Health service provides
free tax-supported healthcare for all residents, including
free access to primary care and all hospitals included in
the present study, and patients with acute conditions such
as PPU are usually referred to the nearest hospital. In
accordance with this, the intervention group was found to
be similar to the three control groups regarding clinical and
demographic characteristics. Patients in the intervention
group tended to smoke less than the historical controls,
but had a greater tendency for alcohol abuse, more co-
morbidity and higher ASA grades, arguing against any
healthy patient bias.
Despite using the preferred method of handling miss-
ing values (multiple imputation31,32 ), missing values could
have affected the validity of the results. Not all patients
were treated according to all elements of the trial protocol,
which could have had an effect on the internal validity of
the results. On the other hand, external validity was high.
There was a minor group of patients at the seven centres
who were eligible for inclusion in the study but who did
not receive the intervention. However, these patients did
not have a higher 30-day mortality rate than the controls,
arguing against the possibility that less ill patients were
selected for intervention. Although the full comparability
of intervention and control groups in this non-randomized
study cannot be assured, the large difference in mortality
rate between the intervention and control groups seems
convincing.
Multimodal perioperative care protocols and fast-track
surgery have been shown to improve outcomes in a number
of surgical specialties, including abdominal surgery13,39 – 43 .
In the present trial, the application of a large number of
www.bjs.co.uk British Journal of Surgery 2011; 98: 802–810
808
evidence-based principles of good clinical practice, includ-
ing those suggested in the Surviving Sepsis Campaign,
was associated with a convincing reduction in the high
mortality rates usually associated with PPU2 – 5,25,44 – 49 .
A core element of the present perioperative care
protocol was the use of diagnostic procedures and
treatment inspired by the Surviving Sepsis Campaign16 ,
including screening for sepsis, initial circulatory and
respiratory stabilization, early use of broad-spectrum
empirical antibiotics, admission to a high-dependency unit,
early goal-directed fluid therapy, and thorough and invasive
monitoring of vital parameters. The mortality rate of septic
shock is up to 50 per cent16,20,50 . Sepsis occurs in up to
20 per cent of patients with PPU, and septic shock may
account for a large proportion of fatalities5,11,12 . Some
25·6 per cent of patients in the intervention group were
diagnosed with septic shock within 30 days of surgery,
compared with 18 per cent in a previous study49 . The
focus on sepsis in the present trial protocol might
explain the seemingly higher proportion of patients with
septic shock in the intervention group. As a result of
the structured and frequent screening for sepsis, more
patients were probably diagnosed with septic shock and
treated accordingly. Similarly, the number of patients
diagnosed with pneumonia in the intervention group
(23·1 per cent) was higher than reported previously49 .
This could be explained by the increased attention to
postoperative complications and frequent (and invasive)
monitoring of vital parameters. However, the increased
number of patients with pneumonia could also be explained
by the relatively liberal fluid therapy (early goal-directed
fluid therapy) and/or the higher proportion of patients
admitted to the ICU, where the prevalence of nosocomial
pneumonia is higher51 . A true increase in the number
of patients with pneumonia and septic shock in the
intervention group seems unlikely, and conflicts with the
significantly improved survival. The increased number of
patients admitted to the ICU, and the longer hospital
stay compared with that in the control groups and a
previous study49 , were probably due to the increased
number of critically ill patients diagnosed with septic shock,
pneumonia and other postoperative complications. The
high frequency of patients admitted to the ICU emphasizes
the optimized perioperative care that was the essence of
the trial.
Based on these results, the authors recommend imple-
menting a perioperative care protocol in patients treated
surgically for PPU. However, it is not possible from the
present findings to determine which elements of the trial
protocol resulted in the improved survival, or whether the
improvement in 30-day mortality was exclusively due to
 2011 British Journal of Surgery Society Ltd
Published by John Wiley & Sons Ltd
M. H. Møller, S. Adamsen, R. W. Thomsen and A. M. Møller
the increased focus and attention to this group of patients
(Hawthorne effect). To answer this clinical question, the
elements of the trial protocol would have to be examined
individually in randomized clinical trials.
Contributors
The following members of the PULP trial group con-
tributed to this study, and are co-authors of this article:
T. Waldau, E. G. Hansen, J. J. Krintel (Copenhagen
University Hospital Herlev, Herlev), L. N. Jørgensen,
M. Rasmussen, B. Belhage (Copenhagen University Hos-
pital Bispebjerg, Copenhagen), J. Bendix, A. Lauritsen
(Aarhus University Hospital, Aarhus), I. Rosenberg, A.
U. Neuenschwander (Copenhagen University Hospital
Hillerød, Hillerød), U. Sigild, A.-M. Hedengrad (Copen-
hagen University Hospital Helsingør, Helsingør), A. G.
Jensen, P. Holde, U. Kelly (Odense University Hospital,
Odense), E. Zimmermann-Nielsen, J. Fristrup (Kolding
Hospital, Kolding).
Acknowledgements
The authors acknowledge the following heads of depart-
ment in the seven participating hospitals: B. Chræmmer,
B. Kyst, F. Moesgaard, K. Ravn, C. Boe Pedersen, G.
Søgaard, H. Harling, J. Glindvad, K. Antonsen, K. Maj-
gaard, B. Hempel Sparsøe, J. Funch Kofoed, P. Klint
Andersen, I. Moritz Hansen, T. Bremholm Rasmussen, H.
Rytter Hansen, J. Budtz, L. Winther Jensen, L. S. Jensen,
M. Jakobsen, L. Kirkegaard, H. Vibeke Andersen, H. K.
Antonsen and C. Jørgensen. They also thank all clinical
staff in the departments of anaesthesiology and surgery in
these hospitals.
Special thanks are due to P. Bartels, B. Randrup Krogh
and A.-M. Sigsgaard Hansen at the Danish National
Indicator Project for access to the database on PPU; and
to D. Bendixen, L. Kure, E.-M. Jacobsen and D. Oxholm
for helping to develop, initiate and run the trial.
The trial was supported financially by the Research
Council, Copenhagen University Hospital Herlev; Depart-
ment of Anaesthesiology and Intensive Care Medicine,
Copenhagen University Hospital Herlev; Aase and Ejnar
Danielsens Foundation; ‘Lægernes Forsikringsforening af
1891’; the Beckett Foundation; the A. P. Møller and Hus-
tru Chastine Mc-Kinney Møllers Foundation; Dr Fritz
Karner and wife Edith Karner’s Foundation; the Tryg
Foundation; and the Danish Society of Anaesthesiology
and Intensive Care Medicine’s Research Foundation. The
authors declare no conflict of interest.
www.bjs.co.uk British Journal of Surgery 2011; 98: 802–810
Optimized perioperative care in perforated peptic ulcer
References
1 Vaira D, Menegatti M, Miglioli M. What is the role of
Helicobacter pylori in complicated ulcer disease?
Gastroenterology 1997; 113(Suppl): S78–S84.
2 Irvin TT. Mortality and perforated peptic ulcer: a case for
risk stratification in elderly patients. Br J Surg 1989; 76:
215–218.
3 Blomgren LG. Perforated peptic ulcer: long-term results
after simple closure in the elderly. World J Surg 1997; 21:
412–414.
4 Thomsen RW, Riis A, Christensen S, Nørgaard M,
Sørensen HT. Diabetes and 30-day mortality from peptic
ulcer bleeding and perforation: a Danish population-based
cohort study. Diabetes Care 2006; 29: 805–810.
5 Møller MH, Adamsen S, Wøjdemann M, Møller AM.
Perforated peptic ulcer: how to improve outcome? Scand J
Gastroenterol 2009; 44: 15–22.
6 Hermansson M, Staël von Holstein C, Zilling T. Surgical
approach and prognostic factors after peptic ulcer
perforation. Eur J Surg 1999; 165: 566–572.
7 Møller MH, Adamsen S, Thomsen RW, Møller AM.
Preoperative prognostic factors for mortality in peptic ulcer
perforation – a systematic review. Scand J Gastroenterol 2010;
45: 785–805.
8 Svanes C, Salvesen H, Espehaug B, Søreide O, Svanes K. A
multifactorial analysis of factors related to lethality after
treatment of perforated gastroduodenal ulcer. 1935–1985.
Ann Surg 1989; 209: 418–423.
9 Egberts JH, Summa B, Schulz U, Schafmayer C, Hinz S,
Tepel J. Impact of preoperative physiological risk profile on
postoperative morbidity and mortality after emergency
operation of complicated peptic ulcer disease. World J Surg
2007; 31: 1449–1457.
10 Koç M, Yoldaş O, Kiliç YA, Göçmen E, Ertan T, Dizen H
et al. Comparison and validation of scoring systems in a
cohort of patients treated for perforated peptic ulcer.
Langenbecks Arch Surg 2007; 392: 581–585.
11 Boey J, Wong J, Ong GB. Bacteria and septic complications
in patients with perforated duodenal ulcers. Am J Surg 1982;
143: 635–639.
12 Fong IW. Septic complications of perforated peptic ulcer.
Can J Surg 1983; 26: 370–372.
13 Kehlet H, Wilmore DW. Evidence-based surgical care and
the evolution of fast-track surgery. Ann Surg 2008; 248:
189–198.
14 Kehlet H. Fast-track colorectal surgery. Lancet 2008; 371:
791–793.
15 Wilmore DW, Kehlet H. Management of patients in fast
track surgery. BMJ 2001; 322: 473–476.
16 Dellinger RP, Levy MM, Carlet JM, Bion J, Parker MM,
Jaeschke R et al. Surviving Sepsis Campaign: international
guidelines for management of severe sepsis and septic shock:
2008. Crit Care Med 2008; 36: 296–327.
17 Moher D, Schulz KF, Altman DG. The CONSORT
statement: revised recommendations for improving the
 2011 British Journal of Surgery Society Ltd
Published by John Wiley & Sons Ltd
809
quality of reports of parallel-group randomized trials. Ann
Intern Med 2001; 134: 657–662.
18 Altman DG, Schulz KF, Moher D, Egger M, Davidoff F,
Elbourne D et al. The revised CONSORT statement for
reporting randomized trials: explanation and elaboration.
Ann Intern Med 2001; 134: 663–694.
19 Centers for Disease Control and Prevention. CDC Guidelines:
Criteria for Defining Nosocomial Pneumonia, 2005;
http://www.vap.kchealthcare.com/vap.aspx [accessed 10
February 2011].
20 Levy MM, Fink MP, Marshall JC, Abraham E, Angus D,
Cook D et al.; SCCM/ESICM/ACCP/ATS/SIS. 2001
SCCM/ESICM/ACCP/ATS/SIS International Sepsis
Definitions Conference. Crit Care Med 2003; 31: 1250–1256.
21 Olsen AH, Njor SH, Vejborg I, Schwartz W, Dalgaard P,
Jensen MB et al. Breast cancer mortality in Copenhagen after
introduction of mammography screening: cohort study. BMJ
2005; 330: 220.
22 Nielsen KA, Jensen NC, Jensen CM, Thomsen M,
Pedersen L, Johnsen SP et al. Quality of care and 30 day
mortality among patients with hip fractures: a nationwide
cohort study. BMC Health Serv Res 2009; 9: 186.
23 Adamsen S, Bendix J, Kallehave F, Wille-Jørgensen P,
Moesgaard F, Nakano A et al. Perforated peptic ulcer. A
national audit. Gut 2005; 54(Suppl VII): A46.
24 Chou NH, Mok KT, Chang HT, Liu SI, Tsai CC,
Wang BW et al. Risk factors of mortality in perforated peptic
ulcer. Eur J Surg 2000; 166: 149–153.
25 Kujath P, Schwandner O, Bruch HP. Morbidity and
mortality of perforated peptic gastroduodenal ulcer following
emergency surgery. Langenbecks Arch Surg 2002; 387:
298–302.
26 Makela JT, Kiviniemi H, Ohtonen P, Laitinen SO. Factors
that predict morbidity and mortality in patients with
perforated peptic ulcers. Eur J Surg 2002; 168: 446–451.
27 Barut I, Tarhan OR, Cerci C, Karaguzel N, Akdeniz Y,
Bulbul M. Prognostic factors of peptic ulcer perforation.
Saudi Med J 2005; 26: 1255–1259.
28 Kocer B, Surmeli S, Solak C, Unal B, Bozkurt B, Yildirim O
et al. Factors affecting mortality and morbidity in patients
with peptic ulcer perforation. J Gastroenterol Hepatol 2007;
22: 565–570.
29 Lee FY, Leung KL, Lai BS, Ng SS, Dexter S, Lau WY.
Predicting mortality and morbidity of patients operated on
for perforated peptic ulcers. Arch Surg 2001; 136: 90–94.
30 Hermansson M, Staël von Holstein C, Zilling T. Surgical
approach and prognostic factors after peptic ulcer
perforation. Eur J Surg 1999; 165: 566–572.
31 Schafer JL. Multiple imputation: a primer. Stat Methods Med
Res 1999; 8: 3–15.
32 Schafer JL, Graham JW. Missing data: our view of the state
of the art. Psychol Methods 2002; 7: 147–177.
33 Knaus WA, Wagner DP, Draper EA, Zimmerman JE,
Bergner M, Bastos PG et al. The APACHE III prognostic
system. Risk prediction of hospital mortality for critically ill
hospitalized adults. Chest 1991; 100: 1619–1636.
www.bjs.co.uk British Journal of Surgery 2011; 98: 802–810
810 M. H. Møller, S. Adamsen, R. W. Thomsen and A. M. Møller

34 Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A,
Knoblich B et al.; Early Goal-Directed Therapy
Collaborative Group. Early goal-directed therapy in the
treatment of severe sepsis and septic shock. N Engl J Med
2001; 345: 1368–1377.
35 Aneman A, Parr M. Medical emergency teams: a role for
expanding intensive care? Acta Anaesthesiol Scand 2006; 50:
1255–1265.
36 Sacks H, Chalmers TC, Smith H Jr. Randomized versus
historical controls for clinical trials. Am J Med 1982; 72:
233–240.
37 Fletcher RH, Fletcher SW. Clinical Epidemiology: The
Essentials (4th edn). Lippincott Williams and Wilkins:
Baltimore, 2005.
38 US Food and Drug Administration. Clinical
Trials Guidance Documents. 2010; http://www.fda.gov/
RegulatoryInformation/Guidances/ucm122046.htm
[accessed 7 February 2011].
39 Pedersen SJ, Borgbjerg FM, Schousboe B, Pedersen BD,
Jorgensen HL, Duus BR et al.; Hip Fracture Group of
Bispebjerg Hospital. A comprehensive hip fracture program
reduces complication rates and mortality. J Am Geriatr Soc
2008; 56: 1831–1838.
40 Møller C, Kehlet H, Friland SG, Schouenborg LO, Lund C,
Ottesen B. Fast track hysterectomy. Eur J Obstet Gynecol
Reprod Biol 2001; 98: 18–22.
41 Marx C, Rasmussen T, Jakobsen DH, Ottosen C,
Lundvall L, Ottesen B et al. The effect of accelerated
rehabilitation on recovery after surgery for ovarian
malignancy. Acta Obstet Gynecol Scand 2006; 85:
488–492.
42 Kehlet H, Wilmore DW. Fast-track surgery. Br J Surg 2005;
92: 3–4.
43 Kehlet H, Wilmore DW. Multimodal strategies to improve
surgical outcome. Am J Surg 2002; 183: 630–641.
44 Svanes C, Salvesen H, Stangeland L, Svanes K, Søreide O.
Perforated peptic ulcer over 56 years. Time trends in
patients and disease characteristics. Gut 1993; 34:
1666–1671.
45 Svanes C. Trends in perforated peptic ulcer: incidence,
etiology, treatment, and prognosis. World J Surg 2000; 24:
277–283.
46 Christiansen C, Christensen S, Riis A, Thomsen RW,
Johnsen SP, Tønnesen E et al. Antipsychotic drugs and
short-term mortality after peptic ulcer perforation: a
population-based cohort study. Aliment Pharmacol Ther 2008;
28: 895–902.
47 Tørring ML, Riis A, Christensen S, Thomsen RW, Jepsen P,
Søndergaard J et al. Perforated peptic ulcer and short-term
mortality among tramadol users. Br J Clin Pharmacol 2008;
65: 565–572.
48 Christensen S, Riis A, Nørgaard M, Sørensen HT,
Thomsen RW. Short-term mortality after perforated or
bleeding peptic ulcer among elderly patients: a
population-based cohort study. BMC Geriatr 2007; 7: 8.
49 Møller MH, Shah K, Bendix J, Jensen AG,
Zimmermann-Nielsen E, Adamsen S et al. Risk factors in
patients surgically treated for peptic ulcer perforation. Scand
J Gastroenterol 2009; 44: 145–152.
50 Dellinger RP, Carlet JM, Gerlach H, Ramsey G, Levy M.
The surviving sepsis guidelines: not another ‘groundhog day’.
Crit Care Med 2004; 32: 1601–1602.
51 Liberati A, D’Amico R, Pifferi S, Torri V, Brazzi L,
Parmelli E. Antibiotic prophylaxis to reduce respiratory tract
infections and mortality in adults receiving intensive care.
Cochrane Database Syst Rev 2009; (4)CD000022.

If you wish to comment on this, or any other article published in the BJS, please
visit the on-line correspondence section of the website (www.bjs.co.uk). Electronic
communications will be reviewed by the Correspondence Editor and a selection
will appear in the correspondence section of the Journal. Time taken to produce
a thoughtful and well written letter will improve the chances of publication in the
Journal.

 2011 British Journal of Surgery Society Ltd www.bjs.co.uk British Journal of Surgery 2011; 98: 802–810
Published by John Wiley & Sons Ltd