Alimentary Pharmacology & Therapeutics
Early versus late surgery for ileo-caecal Crohn’s disease
A. ARATARI*, C. PAPI , G. LEANDROà, A. VISCIDO*, L. CAPURSO  & R. CAPRILLI*
*Gastroenterology Unit, Department of
Clinical Sciences, University of Rome
‘La Sapienza’, Rome, Italy;
 Gastroenterology Unit, S. Filippo
Neri Hospital, Rome, Italy;
àGastroenterology Unit, IRCCS De
Bellis, Castellana Grotte (Bari), Italy
Correspondence to:
Dr A. Aratari, GI Unit, Department of
Clinical Sciences, University of Rome
‘La Sapienza’, Policlinico Umberto I,
Viale del Policlinico 155, 00161
Roma, Italy.
E-mail: aalisa@tin.it
Publication data
Submitted 01 March 2007
First decision 13 April 2007
Resubmitted 09 July 2007
Second decision 16 August 2007
Resubmitted 03 September 2007
Third decision 03 September 2007
Resubmitted 05 September 2007
Accepted 05 September 2007
ª 2007 The Authors
Journal compilation ª 2007 Blackwell Publishing Ltd
doi:10.1111/j.1365-2036.2007.03515.x
SUMMARY
Background
Surgical resection is almost inevitable in Crohn’s disease. Surgery is
usually performed for refractory or complicated disease: no studies
appear to have been carried out, so far, to evaluate the potential bene-
fits of performing surgery early in the course of the disease.
Aim
To compare the long-term course of Crohn’s disease following ileo-cae-
cal resection performed at the time of diagnosis (early surgery) or dur-
ing the course of the disease (late surgery).
Patients and methods
Overall 207 patients with ileo-caecal Crohn’s disease at their first resec-
tion were reviewed: 83 patients underwent surgery at the time of diag-
nosis (early surgery), while 124 underwent surgery 54.2 months (range
1–438) after diagnosis (late surgery). The mean follow-up after surgery
was 147 months (range 12–534). The primary endpoint was clinical
recurrence, defined as need for corticosteroids for symptomatic disease
in the presence of endoscopic and ⁄ or radiologic recurrence. Secondary
endpoints were need for immunosuppressants and surgical recurrence.
Statistical analysis: Kaplan–Meier survival method and Cox proportional
hazards regression model.
Results
Within 10 years after surgery, the cumulative probability of clinical
recurrence was significantly lower in the early surgery group (Log Rank
test P = 0.01). A trend was observed regarding the need for immunosup-
pressants (P = 0.05). No difference was observed regarding surgical recur-
rence. At multivariate analysis, early surgery was the only independent
variable associated with a reduced risk of clinical recurrence (Hazard
ratio, HR = 0.57; 95% CI 0.35 to 0.92, P = 0.02), but not with need for
immunosuppressants and surgical recurrence (HR = 0.51; 95% CI 0.20 to
1.30, P = 0.15; HR = 0.66; 95% CI 0.33 to 1.35, P = 0.25, respectively).
Conclusion
Early surgery prolongs clinical remission compared to surgery per-
formed during the course of the disease, but the natural history of dis-
ease is not modified.
Aliment Pharmacol Ther 26, 1303–1312
1303
1304 A . A R A T A R I et al.
INTRODUCTION
Surgical resection is an almost inevitable event
through the course of Crohn’s disease (CD) and the
surgical rate increases with time. The probability of
surgery is 20–40% during the first year of the disease
and ranges between 30% and 70% 10 years after diag-
nosis; after 15 years, the probability of surgery is as
high as 70–90%.1–6 Surgery is usually performed for
obstructive and septic complications and ⁄ or on
account of refractory disease.7 Several studies have
shown that the clinical behaviour of CD evolves with
time towards obstructive and penetrating complica-
tions, thus the vast majority of patients will ultimately
require surgery.8–10
Surgery is not curative: post-operative recurrence
follows a predictable sequential course: 1 year after
resection, almost 70% of patients have new lesions at
the neoterminal ileum (endoscopic recurrence), and
after 5 years 20–60% will develop symptoms (clinical
recurrence) and 15–50% will need further intestinal
resection for complications or refractory disease (sur-
gical recurrence).3, 11–13 Post-operative prophylactic
treatment with mesalamine is of limited benefit in pre-
venting recurrence.14–16
Although surgery is not curative, in a population
based study, it has been estimated that a representative
CD patient spends 41% of his or her lifetime disease
course in post-surgical remission.17 According to this
observation, surgery appears to be the most effective
therapeutic intervention for inducing sustained remis-
sion. Since 1970, some authors have advocated early
surgery as the approach of choice in CD.18–21 Early sur-
gery may offer several advantages: the re-operation
rate for recurrent disease is not increased,22–24 the
quality of life of the patient improves greatly compared
to prolonged disease activity and long-term medical
treatment,25–27 and early resection, prior to the devel-
opment of advanced or complicated disease, reduces
post-operative morbidity and, possibly, the extent of
the resected specimen.20, 28, 29 In more recent years, the
improvements in medical treatment, particularly the
widespread use of immunosuppressants, have favoured
a conservative approach in the management of CD,
reserving surgery only for refractory or complicated
disease.24, 30, 31 However, despite the widespread use of
immunosuppressants, the natural history of the disease,
in terms of reduction of complications and need for
surgery remains unchanged.32
A prospective trial comparing the long-term benefit
of early surgery versus conventional medical treatment
is not only lacking but is also difficult to perform.
However, in some patients, CD is first diagnosed at
laparotomy carried out on account of the acute or
complicated presentation of the disease. These patients
often undergo surgical resection at the time of diagno-
sis without receiving any specific medical treatment
for CD. This subgroup of patients could be considered
a surrogate model of ‘early surgery’.
The aim of the present study was therefore to assess
the long-term post-operative course in CD patients
undergoing ‘early surgery’ compared to patients
requiring surgical resection during the course of the
disease (late surgery).
PATIENTS AND METHODS
The records of all patients with established diagnosis
of CD referring to our two GI Units in Rome were ret-
rospectively reviewed. Patients who had undergone at
least one radical surgical resection for ileal disease
(with or without right colon involvement) were
included in the study. Patients were divided into two
groups according to the timing of surgery:
– Group 1. (‘early surgery’ or naı̈ve group): patients
in whom surgery was performed at the time of diagno-
sis. These patients underwent surgery for acute or sub-
acute presentation of CD. In this group, the diagnosis
of CD was established at laparotomy and confirmed by
the histo-pathological examination of the resected
specimen. None of these patients had received specific
medical treatment prior to surgery.
– Group 2. (late surgery): patients with established
diagnosis of CD who underwent surgery during the
course of the disease on account of intestinal compli-
cations or refractoriness to medical therapy.
Clinical and demographic characteristics of all
patients were recorded: age at the time of surgery,
sex, smoking habits, family history of inflammatory
bowel disease (IBD), pattern of CD according to the
surgical specimen (penetrating ⁄ non-penetrating), oper-
ative and peri-operative morbidity and prophylactic
treatment of post-operative recurrence. The post-oper-
ative course of all patients was retrospectively evalu-
ated. The primary endpoint was clinical recurrence,
defined as need for systemic corticosteroids for symp-
tomatic disease in the presence of endoscopic and ⁄ or
radiologic recurrence. Secondary endpoints were need
ª 2007 The Authors, Aliment Pharmacol Ther 26, 1303–1312
Journal compilation ª 2007 Blackwell Publishing Ltd
 E A R L Y V S L A T E S U R G E R Y I N C R O H N ’ S D I S E A S E 1305

for immunosuppressants, indicating steroid-dependent ⁄
refractory disease, and surgical recurrence (re-opera-
tion for refractory disease or complications).
Statistical analysis
The Kaplan–Meier survival method was used to esti-
mate the cumulative probability of a post-operative
course free from clinical recurrence, need for immuno-
suppressants and surgical recurrence. Differences
between curves were tested using the Log-Rank test.
Cox proportional hazards regression model was used
to assess whether clinical variables, at the time of the
first resection, were independently associated with the
risk of clinical recurrence, need for immunosuppres-
sants and surgical recurrence during the post-operative
course. The maximum partial likelihood ratio test was
used when entering or removing variables. The Wald
test was used for significance. The proportionality
assumption was graphically tested by plotting {ln [-ln
(survival function)]} with time. The following covari-
ates were considered: age, sex, smoking habits, family
history of IBD, surgical specimen features (penetrating
or non-penetrating), timing of surgery (early surgery
or late surgery) and post-operative prophylactic treat-
ment with mesalamine. BioMedical Data Processing
(BMDP dynamic version 7, University of California,
Los Angeles, CA, USA) was used for all calculations. A
P value <0.05 was considered statistically significant.
Chi-square and Mann–Whitney Rank-Sum test were
used when appropriate.
RESULTS
The records of 605 CD patients were reviewed. Of
these, 207 (34%) underwent at least one surgical resec-
tion for ileal disease with or without right colon
involvement and were included in the study. In 83
patients (group 1, ‘early surgery’), resection was per-
formed for acute or subacute presentation of CD: the
diagnosis of CD had been made at laparotomy and
none of these patients had received specific medical
therapy prior to surgery. In 124 patients (group 2, late
surgery), surgery was performed during the course of
the disease for refractoriness or complications. In
group 2, the mean duration of disease (from diagnosis
to surgery) was 54.2 months (range 1–438). All
patients had received specific CD treatment before sur-
gery: 16 (13%) had received only mesalamine and ⁄ or
antibiotics; 86 (69%) had received at least one course
of systemic corticosteroids and 16 (13%) had received
immunosuppressants.
No operative or peri-operative mortality occurred.
Overall, peri-operative morbidity occurred in 28 of
207 patients (13.5%); major morbidity occurred in 12
of 207 patients (5.8%): eight patients had anastomotic
leak requiring re-operation and temporary stoma, two
patients had intra-abdominal haemorrhage requiring
surgical drainage, one patient presented a severe pan-
creatitis requiring necrosectomy and one patient
underwent early re-operation for ileal volvulus.
The clinical characteristics of the two groups are
shown in Table 1. No statistical difference was
observed with regard to sex, age at surgery, smoking
habits, family history for IBD, pattern of CD at surgical
specimen (penetrating ⁄ non-penetrating), overall peri-
operative morbidity and major peri-operative morbid-
ity. Compared with group 2, patients in group 1 had
less frequently received post-operative treatment with
mesalamine for prevention of recurrence (55% vs.
79%, P = 0.0005). Mean follow-up after surgical resec-
tion was 147 months (range 12–534).

Table 1. Clinical characteristics of patients

Group 1 n. 83 Group 2 n. 124 OR (95%CI) P

Age at surgery (years) mean (range) 36 (12–78) 36 (14–68) ⁄ ns

Sex (F ⁄ M) 35 ⁄ 48 (42% ⁄ 58%) 53 ⁄ 71 (43% ⁄ 57%) 0.97(0.55–1.71) ns
Smoking (yes ⁄ no) 50 ⁄ 33 (60% ⁄ 40%) 73 ⁄ 51 (60% ⁄ 40%) 1.05(0.60–1.86) ns
Family history for IBD (yes ⁄ no) 6 ⁄ 77 (7% ⁄ 93%) 8 ⁄ 116 (6% ⁄ 94%) 1.13(0.37–3.38) ns
Penetrating ⁄ non-penetrating (surgical specimen) 34 ⁄ 49 (41% ⁄ 59%) 60 ⁄ 64 (48% ⁄ 52%) 0.74(0.42–1.29) ns
Peri-operative morbidity n (%) 11 (5.3) 17 (8.2) 0.96(0.43–2.16) ns
Major perioperative morbidity n (%) 5 (6.0) 7 (5.6) 1.07(0.33–0.51) ns
Mesalamine prophylactic treatment (yes ⁄ no) 46 ⁄ 37 (55% ⁄ 45%) 98 ⁄ 26 (79% ⁄ 21%) 0.32(0.17–0.60) 0.0005

IBD, inflammatory bowel disease.

ª 2007 The Authors, Aliment Pharmacol Ther 26, 1303–1312

Journal compilation ª 2007 Blackwell Publishing Ltd
1306 A . A R A T A R I et al.
Post-operative course
Within 120 months after surgery, 83 out of 207 patients
(40.1%) presented clinical recurrence, 25 (12.1%)
required immunosuppressive treatment and 34 (16.4%)
had surgical recurrence. The cumulative probability of a
post-operative course free from clinical recurrence, need
for immunosuppressants and surgical recurrence is
shown in Figure 1. The cumulative probability of a post-
operative course free of clinical recurrence was 77.9%,
68.1%, 56.8%, 48.2% after 30, 60, 90, 120 months,
respectively. The cumulative probability of a course not
requiring immunosuppressants was 94.2%, 89.4%,
87.3%, 83.8% after 30, 60, 90, 120 months, respectively.
Finally the cumulative probability of a course free from
surgical recurrence was 96.7%, 96.1%, 84.5%, 72.1%
after 30, 60, 90, 120 months, respectively.
The cumulative probability of a post-operative
course free from clinical recurrence, need for immuno-
suppressants and surgical recurrence, in groups 1 and
2, is shown in Figure 2. Compared with ‘early surgery’,
resection performed during the course of the disease
was associated with a shorter post-operative course
free of clinical recurrence and need for immunosup-
pressants (log rank test P = 0.01 and P = 0.05, respec-
tively) but no difference between the two groups was
observed regarding the probability of surgical recur-
rence (log rank test P = 0.53). No difference between
the two groups was observed concerning the indica-
tion of re-operation (penetrating vs. non-penetrating
disease) (P = 0.37).
At univariate analysis, none of the other variables
showed any significant difference with regard to the
probability of a post-operative course free of clinical
1.00
Need of immunosuppressants
0.75
Surgical recurrence
0.50
Clinical recurrence
0.25
0.00
0 30 60 90
Patients at risk
Clinical recurrence 207 139 96 71
Need of ISS 207 168 132 112
Surgical recurrence 207 174 134 100
recurrence, need for immunosuppressants and surgical
recurrence. However, patients receiving post-operative
prophylactic treatment with mesalamine had a shorter
survival time free from immunosuppressants (log rank
test P = 0.04) (Table 2).
At multivariate analysis, ‘early surgery’ was the only
independent variable significantly associated with a
reduced probability of clinical recurrence (Hazard
Ratio, HR = 0.57; 95% CI 0.35 to 0.92, P = 0.02), but
not need for immunosuppressants and surgical recur-
rence (HR = 0.51; 95% CI 0.20 to 1.30, P = 0.15;
HR = 0.66; 95% CI 0.33 to 1.35, P = 0.25, respectively)
(Table 3).
A further analysis was then performed on the entire
study population considering early surgery as resec-
tion performed within 6 months of diagnosis (108
patients) and late surgery as resection performed
6 months or more after diagnosis (99 patients). Con-
sidering the same endpoints, results were comparable
(Figure 3).
DISCUSSION
In CD, surgery is usually performed on account of
intestinal complications or disease refractory to medi-
cal treatment, but the timing of surgery is still a
matter of debate.7 The aim of our study was to com-
pare the post-operative course of CD patients follow-
ing ileo-caecal resection performed at the time of
diagnosis (early surgery) or during the course of the
disease (late surgery). We have defined ‘early surgery’
as an operation performed for an acute or subacute
presentation of CD, with laparotomic diagnosis and
subsequent resection. Late surgery has been defined
Figure 1. Cumulative probabil-
ity of a post-operative course
120 Months free from clinical recurrence,
need for immunosuppressants
53
and surgical recurrence in the
94
whole study population.
70
ª 2007 The Authors, Aliment Pharmacol Ther 26, 1303–1312
Journal compilation ª 2007 Blackwell Publishing Ltd
 E A R L Y V S L A T E S U R G E R Y I N C R O H N ’ S D I S E A S E 1307

(a) 1.00

0.75

0.50

0.25

Log rank test P = 0.01

0.00
0 30 60 90 120 Months
Patients at risk
Group1 83 64 50 43 33
Group2 124 76 47 29 20

(b) 1.00

0.75

0.50

0.25

Log rank test P = 0.05

0.00
0 30 60 90 120 Months
Patients at risk
Group1 83 73 63 59 50
Group2 124 96 69 55 44

(c) 1.00

0.75

Figure 2. Cumulative probabil- 0.50

ity of a post-operative course
free from clinical recurrence
(a), need for immunosuppres- 0.25
sants (b), and surgical recur- Log rank test P = 0.53
rence (c) in patients operated
at the time of diagnosis (group 0.00
0 30 60 90 120 Months
1, early surgery ) and dur-
Patients at risk
ing the course of the disease
Group1 83 74 63 49 38
(group 2, late surgery ).
Group2 124 102 72 52 32

as surgery performed during the course of the disease ileal disease (with or without right colon involve-
for complications or refractoriness. A selected popula- ment) submitted to radical resection. The clinical
tion of patients was studied, i.e., only patients with characteristics of the two groups were similar except

ª 2007 The Authors, Aliment Pharmacol Ther 26, 1303–1312

Journal compilation ª 2007 Blackwell Publishing Ltd
1308 A . A R A T A R I et al.

Chi-square for Table 2. Univariate analysis

equivalence P value

Clinical recurrence
Sex F ⁄ M 1.70 0.19
Age at surgical resection (< ⁄ ‡ 40 years) 0.13 0.72
Smoking habit (yes vs. no) 2.92 0.09
Penetrating vs. non-penetrating 1.43 0.23
Family history (yes vs. no) 0.70 0.40
Prophylactic treatment (yes vs. no) 0.31 0.57
Need for immunosuppressants
Sex F ⁄ M 0.29 0.59
Age at surgical resection (< ⁄ ‡ 40 years) 0.01 0.97
Smoking habit (yes vs. no) 0.23 0.63
Penetrating vs. non-penetrating 0.23 0.63
Family history (yes vs. no) 0.56 0.45
Prophylactic treatment (yes vs. no) 4.27 0.04
Surgical recurrence
Sex F ⁄ M 0.02 0.88
Age at surgical resection (< ⁄ ‡ 40 years) 0.54 0.46
Smoking habit (yes vs. no) 2.65 0.10
Penetrating vs. non-penetrating 2.17 0.14
Family history (yes vs. no) 0.13 0.72
Prophylactic treatment (yes vs. no) 3.63 0.06

Covariates HR (95%CI) P value Table 3. Cox proportional

hazards regression
Male 1.27 0.81–2.01 0.30 Clinical recurrence
Age < 40 1.17 0.71–1.92 0.54
Early surgery 0.57 0.35–0.92 0.02
Penetrating 0.81 0.52–1.28 0.36
Smoking 0.82 0.53–1.26 0.36
Family history 1.11 0.50–2.46 0.80
Prophylactic treatment 0.94 0.57–1.56 0.81
Male 1.04 0.47–2.31 0.92 Need for
Age < 40 0.86 0.37–1.97 0.71 immunosuppressants
Early surgery 0.51 0.20–1.30 0.15
Penetrating 1.61 0.71–3.64 0.25
Smoking 1.34 0.57–3.12 0.50
Family history 0.69 0.09–5.15 0.72
Prophylactic treatment 2.44 0.79–7.52 0.12
Male 0.74 0.36–1.50 0.40 Surgical recurrence
Age < 40 1.56 0.68–3.62 0.29
Early surgery 0.66 0.33–1.35 0.25
Penetrating 0.58 0.28–1.21 0.14
Smoking 0.65 0.33–1.29 0.21
Family history 0.50 0.12–2.10 0.34
Prophylactic treatment 0.50 0.24–1.04 0.06

for administration of post-operative prophylactic This can be explained by the fact that most of
mesalamine treatment that was received less fre- these patients underwent surgery on an emergency
quently in patients in the ‘early surgery’ group. basis not in tertiary centres, where post-operative

ª 2007 The Authors, Aliment Pharmacol Ther 26, 1303–1312

Journal compilation ª 2007 Blackwell Publishing Ltd
 E A R L Y V S L A T E S U R G E R Y I N C R O H N ’ S D I S E A S E 1309

(a) 1.00

0.75

0.50

0.25

Log rank test P = 0.02

0.00
0 30 60 90 120 Months
Patients at risk
Group1 108 81 58 48 37
Group2 99 60 39 24 18

(b) 1.00

0.75

0.50

0.25

Log rank test P = 0.03

0.00
0 30 60 90 120 Months
Patients at risk
Group1 108 96 77 65 55
Group2 99 73 55 47 40

(c) 1.00

0.75

Figure 3. Cumulative probabil-

ity of a post-operative course 0.50
free from clinical recurrence
(a), need for immunosuppres-
sants (b), and surgical recur- 0.25
rence (c) in patients who Log rank test P = 0.51
underwent surgery within
6 months of diagnosis ( ) 0.00
and in patients who underwent 0 30 60 90 120 Months
surgery 6 months or more Patients at risk
Group1 108 97 78 55 42
after diagnosis ( ).
Group2 99 78 57 45 30

prophylactic treatment is probably not a common diagnosed. The post-operative course was evaluated
practice. Many of these patients were referred to ter- considering three endpoints: clinical recurrence, need
tiary centres only when symptomatic recurrence was for immunosuppressants and surgical recurrence.

ª 2007 The Authors, Aliment Pharmacol Ther 26, 1303–1312

Journal compilation ª 2007 Blackwell Publishing Ltd
1310 A . A R A T A R I et al.
Clinical recurrence was defined as the need for sys-
temic corticosteroids for the treatment of recurrent
symptoms with documented endoscopic or radiologi-
cal new lesions in the pre-anastomotic ileum. This
arbitrary definition was used on account of the retro-
spective design of our study for which the use of
Crohn’s Disease Activity Index is not suitable.
In the entire study population, the cumulative prob-
ability of clinical recurrence, need for immunosuppres-
sants and surgical recurrence, after 10 years, was 52%,
16% and 28%, respectively. These figures are similar
to those reported by others.3, 12, 33, 34 The most striking
finding emerging from the present study was that in
patients undergoing ‘early surgery’, the post-operative
course free from clinical recurrence, was significantly
longer compared with that in patients in whom sur-
gery was performed late in the course of the disease,
while no statistical difference was observed regarding
surgical recurrence. A trend towards a longer post-
operative course free from immunosuppressants was
also observed. At multivariate analysis, ‘early surgery’
was the only variable independently associated with a
longer post-operative course free from clinical recur-
rence, while the probability of need for immunosup-
pressants and surgical recurrence was not affected.
This appears to indicate that clinical recurrence is
delayed after ‘early surgery’, but the risk of steroid-
dependent ⁄ refractory disease and the development of
intestinal complications requiring further resection
remain unchanged. In other words, ‘early surgery’ pro-
longs surgical-induced remission but, when symptoms
recur, the evolution of the disease is similar to that in
patients submitted to late surgery. This result becomes
even more important, if we consider that patients in
the ‘early surgery’ group were less likely to have been
treated with mesalamine after surgery, compared with
patients in the late surgery group and it is well known
that mesalamine has a little, but significant, effect in
preventing recurrence.15
None of the other variables considered (age, sex,
smoking habit, family history of IBD, penetrating ⁄
non-penetrating disease) seems to affect the post-oper-
ative course. As far as smoking habit is concerned, our
findings are in contrast with data in the literature, in
which smoking is considered a risk factor for clinical
and surgical recurrence.35–37 This disagreement may
be because of the retrospective design of our study.
Data concerning smoking habits were available only at
the time of surgery, but not during the follow-up per-
iod. Thus, it is possible that, during the post-operative
course, many patients changed their smoking habits.
As far as clinical behaviour is concerned, no difference
was observed between penetrating and non-penetrat-
ing CD. Although it is commonly believed that
penetrating disease carries a high risk of early
post-operative recurrence, this is not supported by
consistent evidence.38–40
One possible criticism regarding this study is the
definition of early surgery. Surgery performed at the
first clinical presentation of CD is a ‘surrogate model’
of early surgery. In fact, the typical patient submitted
to early surgery was the one who presented with the
need for urgent surgery and was subsequently found
to have acute ileal CD as the cause of clinical presen-
tation. Therefore, the timing of surgery was deter-
mined by the clinical situation and not related to an
‘extreme top down’ strategy. Conversely, patients in
the ‘late surgery’ group were known to have proven
ileal CD and underwent resection only when conserva-
tive treatment was deemed to have failed. Therefore,
the main characteristics of the ‘early surgery’ group
were a short history of symptoms, an acute or sub-
acute presentation, and no specific medical treatment
before surgery. In our opinion, these characteristics do
not represent a bias in favour of early surgery. In fact,
the pathological findings of the surgical specimen
(penetrating or non-penetrating disease), were similar
in the two groups and there is no consistent evidence
that a short history of symptoms and medical treat-
ment before surgery affect the post-operative course.4
Moreover, our further analysis, defining ‘early surgery’
as surgery performed within 6 months of diagnosis,
gives similar results and provides indirect evidence
that the ‘surrogacy’ is valid. On the other hand, only a
prospective randomized controlled trial could establish
whether early surgery is really beneficial. Such a trial
is, however, very difficult to perform and will proba-
bly never be designed.
The results of the present investigation may have
important clinical implications. According to a conven-
tional ‘step-up strategy’, CD patients with refractory or
steroid-dependent disease are usually candidates to
immunosuppressants and ⁄ or biological treatment
whereas surgery is reserved for non-responders. This
may be a reasonable strategy for patients for whom
surgery is not feasible (high risk patients, extensive
ileal disease, extensive colonic disease), but in patients
with low surgical risk and surgical feasibility (i.e., lim-
ited extent of resection) early surgery should be
strongly recommended. Some authors have reported
ª 2007 The Authors, Aliment Pharmacol Ther 26, 1303–1312
Journal compilation ª 2007 Blackwell Publishing Ltd
 E A R L Y V S L A T E S U R G E R Y I N C R O H N ’ S D I S E A S E 1311

several advantages of early surgery in terms of quality
of life,25–28 and, possibly, lower post-operative morbid-
ity and length of the resected specimen.20, 28, 29 On the
other hand, long-term conservative management with
immunosuppressants has no impact on the need for
surgery or on the rate of intestinal complications.32 In
more recent years, prolonged conservative manage-
ment has led to an increase in elective surgery com-
pared to urgent or emergency surgery but has also led
to an increase in the number of serious complications
before surgery, such as stenosis, subileus, bowel perfo-
rations and peritonitis.31
Furthermore, a Markov analysis has shown that aza-
thioprine treatment and early ileocolonic resection are
both reasonable treatment strategies in patients with
steroid-dependent terminal ileal disease.41 The ECCO
Evidence-Based Consensus on current management of
CD states that ‘surgical options should also be
considered and discussed’ in corticosteroid-dependent
disease.42
In conclusion, as already pointed out, ‘early surgery’
prolongs surgical-induced remission compared to sur-
gery performed late in the course of the disease. The
probability of re-operation is not affected by ‘early
surgery’. Early surgery, therefore, should be carefully
considered as an effective alternative approach to
long-term immunosuppressant treatment in patients
with steroid-refractory or steroid-dependent ileo-cae-
cal CD suitable for radical resection.
ACKNOWLEDGEMENTS
Declaration of personal interests: The authors thank
Mrs Marian Shields for help in reviewing the
English manuscript. Declaration of funding interests:
None.

7 Michelassi F, Balestracci T, Chappel R, 14 Caprilli R, Andreoli A, Capurso L, et al.

REFERENCES
1 Shivananda S, Hordijk ML, Pena AS,
Mayberry JF. Crohn’s disease: risk of
recurrence and re-operation in a defined
population. Gut 1989; 30: 990–5.
2 Hellers G. Crohn’s disease in Stockholm
county 1955–1974. A study of epidemi-
ology, results of surgical treatment and
long-term prognosis. Acta Chir Scand
1979; 490(Suppl.): 1–84.
3 Bernell O, Lapidus A, Hellers G. Risk
factors for surgery and postoperative
recurrence in Crohn’s disease. Ann Surg
2000; 231: 38–45.
4 Munkholm P, Langholz E, Davidsen M,
Binder V. Disease activity courses in a
regional cohort of Crohn’s disease
patients. Scand J Gastroenterol 1995;
30: 699–706.
5 Moum B, Ekbom A, Vatn MH, et al.
Clinical course during the 1st year after
diagnosis in ulcerative colitis and Cro-
hn’s disease. Results of a large, prospec-
tive population-based study in
Southeaster Norway, 1990–93. Scand J
Gastroenterol 1997; 32: 1005–12.
6 Witte J, Shivananda S, Lennard-Jones
JE, et al. Disease outcome in inflamma-
tory bowel disease: mortality, morbidity
and therapeutic management of a 796-
person inception cohort in the European
Collaborative Study on Inflammatory
Bowel Disease (EC-IBD). Scand J Gastro-
enterol 2000; 35: 1272–7.
et al. Primary and recurrent Crohn’s dis-
ease. Experience with 1379 patients.
Ann Surg 1991; 214: 230–40.
8 Louis E, Collard A, Oger AF, Degroote E,
El Yafi FAN, Belaiche J. Behaviour of
Crohn’s disease according to the Vienna
classification: changing pattern over the
course of the disease. Gut 2001; 49:
777–82.
9 Cosnes J, Cattan S, Blain A, et al. Long-
term evolution of disease behavior of
Crohn’s disease. Inflamm Bowel Dis
2002; 8: 244–50.
10 Papi C, Festa V, Fagnani C, et al. Evolu-
tion of clinical behaviour in Crohn’s
disease: predictive factors of penetrating
complications. Digest Liver Dis 2005;
37: 247–53.
11 Rutgeerts P, Geboes K, Vantrappen G,
Kerremans R, Coenegrachts JL, Core-
mans G. Natural history of recurrent
Crohn’s disease at the ileocolonic anas-
tomosis after curative surgery. Gut
1984; 25: 665–72.
12 Borley NR, Mortensen NJ, Jewell DP.
Preventing postoperative recurrence of
Crohn’s disease. Br J Surg 1997; 84:
1493–502.
13 Cottone M, Orlando A, Viscido A, Cal-
abrese E, Camma C, Casa A. Review
article: prevention of postsurgical
relapse and recurrence in Crohn’s dis-
ease. Aliment Pharmacol Ther 2003;
17(Suppl. 2): 38–42.
Oral mesalazine (5-aminosalicylic acid;
Asacol) for the prevention of postopera-
tive recurrence of Crohn’s disease. Ali-
ment Pharmacol Ther 1994; 8: 35–43.
15 Cammà C, Giunta M, Rosselli M, Cot-
tone M. 5-aminosalicylic acid in the
maintenance treatment of Crohn’s dis-
ease: a meta-analysis adjusted for con-
founding variables. Gastroenterology
1997; 113: 1465–73.
16 Cottone M, Cammà C. Mesalamine and
relapse prevention in Crohn’s disease.
Gastroenterology 2000; 119: 597.
17 Silverstein MD, Loftus EV, Sandborn
WJ, et al. Clinical course and costs of
care for Crohn’s disease: Markov model
analysis of a population-based cohort.
Gastroenterology 1999; 117: 49–57.
18 Krause U. Early or late operation in the
treatment of Crohn’s disease. Scand J
Gastroenterol 1971; 6: 479–81.
19 Alexander-Williams J. Surgical aspects
of inflammatory bowel disease. Scand J
Gastroenterol 1990; 172: 39–42.
20 Hulten L. Surgical management and
strategy in classical Crohn’s disease. Int
Surg 1992; 77: 2–8.
21 Lindhagen T, Ekelund GT, Leandor L,
Hildell J, Lindstrom C, Wenckert A. Pre
and Post-operative complications in
Crohn’s disease with special reference to
duration of preoperative disease history.
Scand J Gastroenterol 1984; 19: 194–
203.

ª 2007 The Authors, Aliment Pharmacol Ther 26, 1303–1312

Journal compilation ª 2007 Blackwell Publishing Ltd
1312 A . A R A T A R I et al.
22 Weston LA, Roberts PL, Schoetz DJ Jr,
Coller JA, Murray JJ, Rusin LC. Ileocolic
resection for acute presentation of Cro-
hn’s disease of the ileum. Dis Colon Rec-
tum 1996; 39: 841–6.
23 Andrews HA, Keighley MRB, Alexander-
Williams J, Allan RN. Strategy for man-
agement of distal ileal Crohn’s disease.
Br J Surg 1991; 78: 679–82.
24 Windsor ACJ. Ileal Crohn’s disease is
best treated by surgery. Gut 2002; 51:
11–2.
25 Irvine EJ, Feagan B, Rochon J, et al.
Quality of life: a valid and reliable mea-
sure of therapeutic efficacy in the treat-
ment of inflammatory bowel disease.
Canadian Crohn’s Relapse Prevention
Trial Study Group. Gastroenterology
1994; 106: 287–96.
26 Cohen RD. The quality of life in patients
with Crohn’s disease. Aliment Pharma-
col Ther 2002; 16: 1603–9.
27 Thirlby RC, Land JC, Fenster LF, Lon-
borg R. Effect of surgery on health-
related quality of life in patients with
inflammatory bowel disease: a prospec-
tive study. Arch Surg 1998; 133: 826–
32.
28 Hulten L. Surgical treatment of Crohn’s
disease of the small bowel or ileocecum.
World J Surg 1988; 12: 180–5.
29 Fast S, Helleberg R, Hulten L, Magnus-
son O. Early complications after surgical
treatment for Crohn’s disease with par-
ticular reference to factors affecting
their development. Acta Chir Scand
1980; 146: 519.
30 Scott NA, Hughes LE. Timing of ileoco-
lonic resection for symptomatic Crohn’s
disease – the patient’s view. Gut 1994;
35: 656–7.
31 Siassi M, Weiger A, Hohenberger W,
Kessler H. Change in surgical therapy
for Crohn’s disease over 33 years: a pro-
spective longitudinal study. Int J Colo-
rectal Dis 2007; 22: 319–24.
32 Cosnes J, Nion-Larmurier I, Beaugerie L,
Afchain P, Tiret E, Gendre JP. Impact of
the increasing use of immunosuppres-
sants in Crohn’s disease on the need for
intestinal surgery. Gut 2005; 54: 237–
41.
33 Bernell O, Lapidus A, Hellers G. Risk
factors for surgery and recurrence in
907 patients with primary ileocecal Cro-
hn’s disease. Br J Surg 2000; 87: 1697–
701.
34 Veloso FT, Ferreira JT, Barros L, Alme-
ida S. Clinical outcome of Crohn’s dis-
ease: analysis according to the Vienna
classification and clinical activity. In-
flamm Bowel Dis 2001; 7: 306–13.
35 Cottone M, Roselli M, Orlando A, et al.
Smoking habits and recurrence in Cro-
hn’s disease. Gastroenterology 1994;
106: 643–8.
36 Yamamoto T, Keighley MRB. Smoking
and disease recurrence after operation
ª 2007 The Authors, Aliment Pharmacol Ther 26, 1303–1312
Journal compilation ª 2007 Blackwell Publishing Ltd
for Crohn’s disease. Br J Surg 2000; 87:
398–404.
37 Birrenbach T, Bocker U. Inflammatory
disease and smoking. A review of epide-
miology, pathophysiology and therapeu-
tic implications. Inflamm Bowel Dis
2004; 10: 848–59.
38 Greenstein AJ, Lachman P, Sachar DB,
et al. Perforating and non-perforating
indications for repeated operations in
Crohn’s disease: evidence for two clini-
cal forms. Gut 1988; 29: 588–92.
39 Yamamoto T, Allan RN, Keighley MR.
Perforating ileocecal Crohn’s disease
does not carry a high risk of recurrence
but usually re-presents as perforating
disease. Dis Colon Rectum 1999; 42:
519–24.
40 Yamamoto T. Factors affecting recur-
rence after surgery for Crohn’s disease.
World J Gastroenterol 2005; 11: 3971–
9.
41 Kennedy ED, Urbach DR, Krahn MD,
Steinhart AH, Cohen Z, McLeod RS.
Azathioprine or ileocolonic resection for
steroid-dependent terminal ileal Crohn’s
disease? A Markov analysis. Dis Colon
Rectum 2004; 47: 2120–30.
42 Travis SP, Stange EF, Lémman M, et al.
European evidence-based consensus on
the diagnosis and management of Cro-
hn’s disease: current management. Gut
2006; 55(Suppl. 1): i16–35.