A prospective, longitudinal study of nonconventional strictureplasty in Crohn’s disease1

A Prospective, Longitudinal Study of Nonconventional Strictureplasty in Crohn’s Disease Gianluca M Sampietro, MD, Massimo Cristaldi, MD, FACS, Giovanni Maconi, MD, Fabrizio Parente, MD, Alessandra Sartani, MD, Sandro Ardizzone, MD, Piergiorgio Danelli, MD, Gabriele Bianchi Porro, MD, Angelo Maria Taschieri, MD Bowel-sparing techniques have been proposed to avoid extended or repeated resections in patients with Crohn’s disease (CD), but without precise indications, prospective evaluation, and with a technically limited repertoire. STUDY DESIGN: A prospective longitudinal study of new nonconventional strictureplasties (NCSP) in order to evaluate the safety, type and site of recurrence, and longterm clinical and surgical efficacy. RESULTS: Between January 1993 and December 2002, 102 among 305 consecutive patients underwent at least one NCSP for complicated CD. Patients were treated following precise indications and then included in a prospective database with scheduled followup. Factors claimed to influence postoperative and longterm outcomes and type and site of recurrence were analyzed. We performed 48 ileoileal side-to-side isoperistaltic strictureplasty (SP), 41 widening ileocolic SP, 32 ileocolic side-to-side isoperistaltic SP, associated with Heineke-Mikulicz SP (in 80 procedures) or with minimal bowel resections or both (in 47 procedures). Postoperative mortality was nil; complication rate was 5.7%. Ten years clinical and surgical recurrence rates were 43% and 27%, respectively. Recurrence rate on an NCSP site was 0.8%. No specific factor was identified as related to postoperative or longterm outcomes. CONCLUSIONS: Perioperative and longterm results of NCSP are comparable to or even better than both conservative and resective surgery as reported in the literature, with a low recurrence rate on the NCSP site. Considering the unpredictability of the clinical course of CD and the lifetime need for surgical procedures, NCSP, together with minimal resection and classic SP repertoire, should be considered first-line treatment in complicated CD. ( J Am Coll Surg 2004;199:8–22. © 2004 by the American College of Surgeons) BACKGROUND:

Operation for Crohn’s disease (CD) is a part of the clinical history, considering that within 10 to 20 years from diagnosis the majority of patients undergo at least one surgical intervention.1-3 Some of them undergo extended or repeated resections with consequent different

levels of intestinal malabsorption and the risk of developing short bowel syndrome.4,5 In the past decade, some authors have reported on series of patients treated with Heineke-Mikulicz and Finney strictureplasty (SP) with safe and effective results.6-16 Reports refer mainly to relatively small retrospective series with short followup and scarce information about CD location and behavior. Pre- and postoperative medical therapy has changed widely through the 1970s, 1980s, and 1990s, influencing outcomes. Another major problem is the lack of precise indications, applied prospectively, for the use of SP in most series. Heineke-Mikulicz and Finney SP are not feasible in many situations, such as with the presence of long, multiple, or close jejunoileal strictures; strictures of the ileocecal region; or in cases of previous ileocolonostomy, where bowel preservation would be achievable or preferable. To expand use of bowel-sparing

No competing interests declared.

Presented at the American College of Surgeons 89th Annual Clinical Congress, Chicago, IL, October 2003. Received September 5, 2003; Revised January 27, 2004; Accepted January 28, 2004. From the Department of Surgery, II Division of General Surgery (Sampietro, Cristaldi, Sartani, Danelli, Taschieri) and the Department of Medicine, Division of Gastroenterology (Maconi, Parente, Ardizzone, Porro), Universita` degli Studi di Milano, Dipartimento di Scienze Cliniche “Luigi Sacco,” Ospedale Luigi Sacco, Milan, Italy. Correspondence address: Gianluca M Sampietro, Department of Surgery, II Division of General Surgery, Universita´ degli Studi di Milano, Dipartimento di Scienze Cliniche “Luigi Sacco,” Ospedale Luigi Sacco, Via GB Grassi 74, 20157 Milan, Italy.

© 2004 by the American College of Surgeons Published by Elsevier Inc.

8

ISSN 1072-7515/04/$30.00 doi:10.1016/j.jamcollsurg.2004.01.039

Vol. 199, No. 1, July 2004

Sampietro et al

Nonconventional Strictureplasty in Crohn’s Disease

9

Table 1. Vienna Classification System of Crohn’s Disease21 Abbreviations and Acronyms

CD ⫽ Crohn’s disease NCSP ⫽ nonconventional strictureplasty SP ⫽ strictureplasty

techniques, some authors have recently proposed original procedures for such situations: ileoileal side-to-side isoperistaltic SP,17-19 ileocolic side-to-side isoperistaltic17 or antiperistaltic SP,20 and ileocecal SP.17 The aim of this prospective, longitudinal study was to assess safety and longterm efficacy of nonconventional strictureplasties (NCSP) in a large series of patients treated for complicated CD. In addition, the type and site of CD recurrence and individual variables possibly associated with postoperative outcomes and surgical and clinical recurrence were analyzed. METHODS Three-hundred five patients were treated in our department for primary or reoperative CD between January 1993 and December 2002. Two-hundred fifty-two patients had at least 1 small bowel localization at the time of surgery, 129 underwent 1 or more SPs, and 102 underwent at least 1 NCSP. These patients form the basis of the current study. At hospital admission, all patient data concerning the characteristics of CD (ie, age at diagnosis, location and behavior of the disease); clinical and surgical history; preoperative biochemical, endoscopic, and radiologic findings; and indication for surgery were recorded in a computerized database. We classified our patients using the Vienna Classification system21 and, on the basis of the data collected, the patients treated before 1998 were reclassified according to this method (Table 1). The decision to operate was made on the basis of clinical, endoscopic, radiologic, and ultrasonographic findings. The postoperative followup was recorded by computing data from clinical examination, biochemical tests, ultrasonography, and endoscopy. Visits and tests were performed at 3, 6, and 12 months during the first year; every 6 months for the next 4 years; and then on a yearly basis in case no clinical or surgical recurrence occurred. In patients with incomplete obstruction, a bowel preparation with 1 L 2 days before and 4 L 1 day before surgery with polyetilenglycole solution was used, and prophylaxis was performed with piperacillin and metro-

Characteristic

Age at diagnosis (y) ⬍40 ⱖ40 Location Terminal ileum Colon Ileocolon Upper gastrointestinal Behavior Nonstricturing nonpenetrating Stricturing Penetrating Further data to be collected Patient’s name and date of birth Gender: female/male Ethnicity: Caucasian/African American/ Asian/other Jewish: yes/no/partly Family history of inflammatory bowel disease: first-degree relatives/other/none Extraintestinal manifestations: yes/no

Vienna Classification

A1 A2 L1 L2 L3 L4 B1 B2 B3

nidazole in almost all patients. The operating surgeon completed a special form at the end of each procedure listing the intraoperative findings and surgical techniques used. As described previously, a conservative surgical approach was considered the treatment of choice in all patients operated on by the authors (GMS, MC, PD, AMT).17,22,23 The surgical decision-making algorithm is detailed in Figure 1. Figure 2 presents a schematic diagram of the NCSP techniques designed by the senior author and used in our series. All the intestinal sutures were performed with interrupted stitches for the seromuscolar layer and a row of running sutures for mucosal layer using coated, braided, lactomeric, absorbable 3-0 sutures. In the ileocolic and ileoileal side-to-side isoperistaltic strictureplasties, the anastomotic ends were tapered to avoid blind stumps. In case of bleeding, additional hemostatic stitches should be added on the mucosal side. At least one full-thickness biopsy was obtained from each SP, and biopsies of areas suspicious for malignancy were obtained for frozen section (all negative). In case the surgeon decides on a side-to-side SP type, in the presence of thickened, retracted mesentery with enlarged nodes, the peritoneal layer of the mesentery is not opened before proceeding in vessel ligatures because this maneuver could cause hemorrhage and

10

Sampietro et al

Nonconventional Strictureplasty in Crohn’s Disease

J Am Coll Surg

Figure 1. Surgical decision-making algorithm used in the study.

jeopardize the bowel segment. We prefer treating the mesentery by digitoclasia, holing it using an O’Shougnessy clamp and then tying the whole mesentery before cutting it; transfixed stitches are recommended to complete the hemostasis. At the end of each

procedure abdominal lavagewith at least 2 L of warm saline solution was performed. The study was approved by the local ethics committee and informed consent was obtained from all patients. Clinical recurrence was defined as the presence of

Vol. 199, No. 1, July 2004

Sampietro et al

Figure 2. Line diagram of the surgical procedures side isoperistaltic strictureplasty; 48 procedures ileocolic side-to-side isoperistaltic strictureplasty; Lower panel, widening ileocolic strictureplasty; 41

Nonconventional Strictureplasty in Crohn’s Disease

11

designed by the senior author.17 Upper panel, ileoileal side-toperformed; length: min. 10 cm, max. 60 cm. Middle panel, 32 procedures performed; length: min. 10 cm, max. 25 cm. procedures performed; length: max. 10 cm.

CD-related abdominal symptoms associated with radiologic, endoscopic, and laboratory findings requiring treatment with medium-high doses of steroids. Surgical recurrence was defined as the presence of therapy refractory disease or CD-related complications requiring a new surgical procedure. As part of another ongoing study, the majority of patients were randomly assigned to a postoperative medical treatment with 5-ASA compounds or immunosuppressors. Statistical analysis was performed using chi-square test or Fisher’s exact test for categorical variables and the Mann-Whitney U test for continuous variables. Cumulative probability of clinical and surgical recurrences was estimated by the time-toevent Kaplan and Meier’s method. To assess the relative excess risk of events the Cox’s proportional hazards model was used. Level of statistical significance was set at p ⬍ 0.05.

RESULTS We performed a total of 248 procedures in 102 patients: 48 ileoileal side-to-side isoperistaltic SP, 41 widening ileocolic SP, 32 ileocolic side-to-side isoperistaltic SP, 80 associated Heineke-Mikulicz SP, and 47 associated minimal small bowel resections. The vast majority of patients underwent operation for stenosis as primary indication (ⱖ98%); 2 patients necessitated drainage of an abscess close to the stenotic segment. All patients were Caucasian, without known Jewish origins. Sixty-five patients were men (men:women ratio 1.8:1), 80 patients were diagnosed as having CD before the age of 40 (A1 78.4%, A2 21.6%), mean CD duration was 7.3 ⫾ 5.2 years. Mean age at first and current surgery was 34.2 ⫾ 10.7 and 39.2 ⫾ 11.6 years, respectively. Mean time between diagnosis and first surgery was 3.4 ⫾ 4.2 years, and between last previous and present surgery was 6.6 ⫾ 5.2 years. At hospital admis-

12

Sampietro et al

Nonconventional Strictureplasty in Crohn’s Disease

sion 53 patients were smokers (51.9%), family history of inflammatory bowel disease was positive in 8 patients (7.8%), and extraintestinal manifestations were reported by 13 patients (12.7%). Fifty-four patients (52.9%) had already undergone 75 previous surgical procedures (range 1 to 7). Conservative operation (minimal ileal resection, strictureplasty, or both) accounted for 17.5% of the cases. According to the Vienna Classification, 74 patients had an ileal localization (L1 72.5%), 4 had ileocolic (L3 4%), and 24 had an upper gastrointestinal (L4 23.5%) localization, while 65 patients (63.7%) had stricturing (B2) and 37 (36.3%) penetrating (B3) disease behavior (31.3% had a concomitant fistula or an abscess or both at the time of current operation). Preoperative therapy was based on steroids in 46 patients (45%), 5-ASA compounds in 25 (24.5%), immunosuppressive drugs in 25 (24.5%), and 6 patients (6%) were under clinical observation without therapy. Immunosuppressive drugs were suspended at least 30 days before operation, steroids at different times depending on clinical presentation, but at least 48 hours before operation after tapering. As postoperative adjuvant therapy, 59 patients received 5-ASA compound (57.8%), 37 immunosuppressive drugs (36.2%), and 6 topical steroids (6%). Such therapy was maintained unless side effects or clinical recurrence occurred. Postoperative mortality was nil; anastomotic leak occurred in three patients (2.9%), self-limited luminal bleeding requiring transfusion occurred in two patients (1.9%), and prolonged ileus (⬎7 days) in one patient (0.9%). Anastomotic leak occurred in the following three patients. First was a female patient (patient #34-A1L1B3) who had undergone two previous operations, was a smoker, and had undergone an ileoileal side-to-side and an ileocolic side-to-side SP plus a minimal bowel resection. At reoperation, she presented with 2 SP leakage sites of ⬍1 cm and was treated by direct suture. Second, a male patient (patient #76-A1L1B2) with one previous intervention, current smoker, who had received an ileoileal side-to-side and a widening ileocolic SP was treated conservatively, but developed an adhesion between the NCSP and the sigmoid colon that necessitated a laparotomy for intestinal obstruction after 12 months, a simple adhesiolysis was performed. Third, a male patient (patient #79-A1L1B3), without previous operation, a smoker, treated by an ileocolic side-to-side SP plus a minimal bowel resection, had a wide SP leakage

J Am Coll Surg

and received a minimal bowel resection with lateral ileostomy, taken down 3 months later. Factors considered for possible association with postoperative septic complications, mean ⫾ SD, Fisher’s exact test, or Mann-Whitney U test p values were: age (A1 3 versus 77, p ⫽ 1), gender (2 men versus 63 women, p ⫽ 1), smoking habit (present in 3 versus 50, p ⫽ 0.24), previous operation (present in 2 versus 55, p ⫽ 1), concomitant resection (present in 2 versus 44, p ⫽ 0.58), CD location and behavior (L1 in 3 versus 70, p ⫽ 1; B3 in 2 versus 35, p ⫽ 0.29), preoperative erythrocyte sedimentation rate (33.7 ⫾ 10.8 versus 27.4 ⫾ 20 mm/h, p ⫽ 0.26), C-reactive protein (0.7 ⫾ 0.8 versus 5.3 ⫾ 19.8 mg/dL, p ⫽ 0.97), hemoglobin levels (13.8 ⫾ 2.8 versus 12.8 ⫾ 2.1 g/dL, p ⫽ 0.34), WBC count (8.6 ⫾ 1 versus 8.3 ⫾ 3.6 *103/␮L, p ⫽ 0.41), platelets (297.3 ⫾ 165.7 versus 298.2 ⫾ 94.3 *103/␮L, p ⫽ 0.78), total serum proteins (6.5 ⫾ 0.3 versus 6.5 ⫾ 0.7 g/dL, p ⫽ 0.97), and albumin (3.8 ⫾ 0.3 versus 3.9 ⫾ 0.8 g/dL, p ⫽ 1). During postoperative followup (median 41.2 months, range 1 to 120 months) 15 patients experienced a surgical recurrence and 27 a clinical recurrence, 2 patients were lost to followup and censored at the last clinical observation. The 10-year cumulative clinical and surgical recurrences were 43% and 27%, respectively, and the 10-year prevalence of short bowel syndrome was 0.9%. Recurrence occurred in only 3 of 201 procedures on an SP site (1.5%), and in only 1 of 121 on an NCSP site (0.8%). Table 2 summarizes the data concerning the type and site of surgical recurrence. At reoperation 60% of the patients had SP again, and 50% of those also underwent minimal bowel resection, 27% of the reoperated patients had only minimal bowel resection, and 13% only colonic resection. Clinical and surgical recurrence were not related to demographic, clinical, and biochemical features. Tables 3 and 4 report the comparisons between the variables and Cox’s proportional hazard model results. The Kaplan-Meier time-to-event estimates and the recurrence-free percentages for clinical and surgical recurrence are shown in Figure 3. DISCUSSION Bowel stricture is a frequent intestinal complication of CD occurring at some time during the natural course of the illness and constituting the most common indication for operation in these patients.

Site of recurrence Patient No.

Type of operation

Same intestinal location

New intestinal location

Strictureplasty

Anastomosis

2

A1L4B2

Female

3

A1L1B2

Male

6 8 14 18

A1L1B2 A1L1B3 A2L1B2 A1L1B2

Male Male Male Male

21

A1L4B2

Male

22

A1L1B2

Male

24

A1L4B2

Male

40

A1L1B2

Male

42

A1L4B2

Female

53 60

A1L1B2 A1L4B2

Male Male

66

A1L4B2

70

A1L1B2

Same behavior

New behavior

1 Ileoileal side-to-side 1 Widening ileocolic 2 Heineke-Mikulicz 1 Widening ileocolic 1 Ileocolic side-to-side 1 Ileocolic side-to-side 2 Ileoileal side-to-side 1 Heineke-Mikulicz 1 Ileoileal side-to-side 1 Heineke-Mikulicz 1 Ileocolic side-to-side 1 Heineke-Mikulicz 1 Ileocolic side-to-side 4 Heineke-Mikulicz 1 Ileocolic side-to-side 1 Ileocolic side-to-side

Yes

—

—

Yes Ileocolonostomy —

Yes

—

Yes

—

Yes

—

Yes — Yes Yes

— Yes-L1 3 L3 — —

— — — —

Yes Yes — Yes

— — Yes-B2 3 B3 —

Yes

—

—

—

Yes-B2 3 B3

Yes

—

—

Yes

—

Yes

—

—

—

Yes

—

Yes

—

Yes Ileocolic side-to-side —

—

—

Yes-B2 3 B3

Yes

—

— Yes

Yes-L1 3 L3 —

Female

2 Heineke-Mikulicz 1 Ileoileal side-to-side 1 Widening ileocolic 10 Heineke-Mikulicz 1 Ileoileal side-to-side 1 Ileocolic side-to-side

—

Yes

—

— Yes Heineke-Mikulicz —

— —

— Yes

Yes-B2 3 B3 —

Yes

—

Yes Ileocolonostomy —

Yes

—

Male

1 Ileoileal side-to-side

Yes

—

—

—

Yes Heineke-Mikulicz — — — —

—

Yes Ileoileostomy —

*A1, ⬍40 years; A2, ⱖ40 years; L1, terminal ileum; L2, colon; L3, ileocolon; L4, upper gastrointestinal; B1, nonstricturing nonpenetrating; B2, stricturing; B3, penetrating.

Nonconventional Strictureplasty in Crohn’s Disease

Gender

Sampietro et al

Vienna Classification*

Type of recurrence

Vol. 199, No. 1, July 2004

Table 2. Type and Site of Recurrence in Terms of Behavior and Location—Recurrence on a Previous Strictureplasty or Anastomosis Is Evidenced Separately

13

14

Sampietro et al

Nonconventional Strictureplasty in Crohn’s Disease

J Am Coll Surg

Table 3. Clinical Recurrence—Comparisons of Proportions and Cox Proportional Hazard Model Results

Patients Characteristic

Age (y) A1 (⬍ 40) A2 (ⱖ 40) Gender Male Female Familiarity Present Absent Smoking habit Present Absent Extraintestinal manifestations Present Absent Previous surgery Present Absent Age at first surgery (y) Time to first surgery (y) Time between previous and our surgery (y) CD duration (y) Synchronous resection Present Absent CD location L1 (terminal ileum) L3 (ileocolon) L4 (upper gastrointestinal) CD behavior B2 (stricturing) B3 (penetrating) ESR (normal 1–20 mm/h) CRP (normal ⬍ 1 mg/dL) Therapy pre/post* Steroids 5-ASA compounds Immunusuppressors No therapy

Fisher’s exact, chi-square, or Mann-Whitney U tests

Cox’s proportional hazard model

With clinical recurrence

Without clinical recurrence

p Value

p Value

21 6

59 16

0.99

0.97

18 9

47 28

0.81

0.44

3 24

5 70

0.43

0.12

18 9

35 40

0.1

0.14

4 23

9 66

0.74

0.64

17 10 35.5 ⫾ 11.2 3.1 ⫾ 3.3

40 35 33.6 ⫾ 10.6 3.6 ⫾ 4.5

0.49 0.58 0.97

0.3

6.2 ⫾ 4.6 7.4 ⫾ 4.6

6.8 ⫾ 5.6 7.3 ⫾ 5.4

0.71 0.77

9 18

37 38

0.18

0.47

18 3 6

56 2 17

0.21

0.77

21 6 28.7 ⫾ 25.2 7.5 ⫾ 25.9

44 31 26.4 ⫾ 19.9 4.3 ⫾ 17.1

0.1 0.98 0.85

0.12

12/2 9/12 6/13 0/0

34/4 16/47 19/24 6/0

0.45/0.25

*Postoperative therapy received from the patients until the clinical recurrence. CD, Crohn’s disease; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate.

The current study reports on the most consistent prospective series of NCSPs performed as the first-choice surgical procedure for stenotic CD in a single center over a 10-year period. Resective bowel surgery has been, and

continues to be, the mainstay of surgical treatment for CD, performed in 85% to 90% of primary and reoperative surgery.9,24,25 The possible impact of bowel resection on intestinal absorption, especially for the terminal il-

Sampietro et al

Vol. 199, No. 1, July 2004

Nonconventional Strictureplasty in Crohn’s Disease

15

Table 4. Surgical Recurrence—Comparisons of Proportions and Cox Proportional Hazard Model Results

Patients Characteristic

Age (y) A1 (⬍ 40) A2 (ⱖ 40) Gender Male Female Familiarity Present Absent Smoking habit Present Absent Extraintestinal manifestations Present Absent Previous surgery Present Absent Age at first surgery (y) Time to first surgery (y) Time between previous and our surgery (y) CD duration (y) Synchronous resection Present Absent CD location L1 (terminal ileum) L3 (ileocolon) L4 (upper gastrointestinal) CD behavior B2 (stricturing) B3 (penetrating) ESR (normal 1–20 mm/h) CRP (normal ⬍ 1 mg/dL) Therapy pre/post* Steroids 5-ASA compounds Immunusuppressors No therapy

Fisher’s exact, chi-square, or Mann-Whitney U tests

Cox’s proportional hazard model

p Value

p Value

With surgical recurrence

Without surgical recurrence

12 3

68 19

1

0.89

12 3

53 34

0.24

0.28

1 14

7 80

1

0.61

10 5

43 44

0.26

0.16

2 13

11 76

0.97

0.47

10 5 36.1 ⫾ 11.7 2.3 ⫾ 2.7

47 40 33.7 ⫾ 10.6 3.6 ⫾ 4.4

0.41 0.52 0.31

0.21

5.9 ⫾ 3.9 7.1 ⫾ 4.5

6.8 ⫾ 5.5 7.3 ⫾ 5.3

0.68 0.99

8 7

38 49

0.57

0.06

9 0 6

65 5 17

0.16

0.64

11 4 23.8 ⫾ 18.1 9.8 ⫾ 31.7

54 33 28.5 ⫾ 20.2 4.3 ⫾ 16.1

0.56 0.29 0.78

0.06

5/0 6/9 4/6 0/0

41/6 19/50 25/31 6/0

0.45/0.57

*Postoperative therapy received from the patients until the clinical recurrence. CD, Crohn’s disease; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate.

eum and jejunum, has been poorly investigated and it is difficult to quantify and qualify absorption deficits in daily clinical practice. There is evidence that even simple terminal ileum resection at primary operation causes

folic acid and vitamin B12 deficiencies.26,27 Intestinal malabsorption is a “100-step stair” and it is difficult to define at which level a single Crohn’s patient “stands” and how fast he will “run down.” At the end of the stair lies

16

Sampietro et al

Nonconventional Strictureplasty in Crohn’s Disease

J Am Coll Surg

Figure 3. Kaplan-Meier time-to-event estimates and recurrence-free percentages for clinical (upper panel) and surgical (lower panel) recurrence. (A) Kaplan-Meier function for clinical recurrence. O, complete; ⫹, censored. (B) Kaplan-Meier function for surgical recurrence. O, complete; ⫹, censored.

the short bowel syndrome, which involves 1.5% to 15% of patients, depending on the different published series.4,28-30 These patients are managed by total parenteral nutrition, but this option is not without cost and may cause complications such as hepatic disease or sepsis, or even death. Total parenteral nutrition mortality rates for benign gastrointestinal disorders are reported to be 9% at 1 year, 30% at 3 years, and 38% at 5 years.31 In our series, although 52.9% of the patients underwent one or more

previous surgical procedures, only one patient (0.9%), who underwent seven previous interventions in other centers (patient #52-A1L4B3), had short bowel syndrome. Since their first description, the performance of SPs has been improving, but only as an alternative surgical option for selected groups of patients and without precise indications, in spite of the early encouraging results indicating overall morbidity rates of 5.8% to 18%, SP-

Vol. 199, No. 1, July 2004

Sampietro et al

related complications of 0% to 13%, and mortality rates of 0% to 0.5%.7-17,22,32 Our perioperative results report a morbidity rate of 5.7% and no mortality and are comparable to or even better than those of the major resective and SP series reported in the literature. Septic complications and self-limited intraluminal bleeding, the most common complications after operation for CD,7-17,22,33,34 account for 2.9% and 1.9% of our patients, respectively. In the past, malnutrition, increasing age, weight loss, abdominal sepsis, emergency operation, and low albumin and hemoglobin levels have been proposed as factors affecting the incidence of postoperative complications after SP, but in our series none of these variables reached statistical significance.12,14,22,32,35-38 Our series of patients did not seem to differ from those reported in the literature if one considers the duration of disease (more than 10 years in 32.6%) and the number of patients who underwent previous operations (52.9%, 82.5% of whom were treated by wide resective operation). The reasons for these results are probably factors such as the mean age of our population was quite low (39.2 ⫾ 11.6, A1:78.2%); none of the patients underwent emergency operation for free perforation, toxic dilation, or massive luminal bleeding; no patient underwent operation with serum albumin levels ⬍ 2.5 g/dL; and only 7 patients underwent surgery with hemoglobin levels ⬍ 10 g/dL (min. 8.5 g/dL). In our opinion, the role of preoperative enteral or parenteral therapy, or both, needs to be stressed in order to optimize the timing and benefits of the surgical procedure. Another prognostic factor claimed to influence either the postoperative outcomes or longterm recurrence is the penetrating behavior of the disease.36,37,39-41 In the current series, the B3-behavior was not considered an absolute contraindication to perform the SP (see the surgical decisionmaking algorithm in Fig. 1) and even if 36.3% of patients had been defined as B3, this characteristic did not influence morbidity. This is also true for the two patients who had abscess drainage as primary indication, as well as in the 31.3% with concomitant fistulae or abscesses, or both. Our indications have been applied prospectively, thus reducing the risk of a selection bias by the operating surgeon in the way that the best segments undergo SP while the worst are resected, with consequent better results in favor of SP. Even the disease location has not been considered a contraindication to perform NCSP in this study, and L1, L3, and L4 patients are present in this series. This is

Nonconventional Strictureplasty in Crohn’s Disease

17

somehow progress because the L4-upper-gastrointestinal location is the only one considered worldwide as an indication to perform SP29,30,42 (Heineke-Mikulicz SP only and when feasible), while the last third of the ileum and the ascending colon are usually subjected to resections. In our experience, the pure L2-colonic location is the only intestinal tract not considered feasible for an NCSP procedure, principally because of the anatomic characteristics, vascularization, and disease behavior of this segment and the high potential for cancer. With few exceptions, this is a widespread trend.43 The necessity to perform a concomitant bowel resection, always performed without disease-free margins, has been 18.9% in this study, against 66% to 85% in the major SP series.9,13,14,16,19,30,44,45 Michelassi and colleagues,19 despite the very encouraging results of their recent work, consider intestinal resection the mainstay of treatment for CD, and reserve the side-to-side isoperistaltic SP to only 4.5%, and the use of conventional SP to only 11.5%, of their patients. Dietz and colleagues,14 in the Cleveland Clinic Foundation’s 15 years’ experience of 1,124 SPs, report a concomitant bowel resection in 66% of patients, 75% of whom had diffuse diseased segments and long, multiple, and close strictures (“chain of sausages”), all situations in which the NCSP would be feasible. The very high number of SPs per patient in this study (median 2, range 1 to 19) is because the Heineke-Mikulicz technique is the most extensively used, but the total number of intestinal sutures during surgery is considered a risk for postoperative septic complications, another aspect in favor of NCSP. The only historical SP technique feasible for segments longer than 8 cm or for multiple and close strictures is the Finney SP, used by Dietz and colleagues14 in 80 patients. The consequence of such a procedure is an intestinal bypass with a giant lateral diverticulum, resulting in early recurrence, a tendency shown also in Dietz’s patients, and the potential for cancer. Using our side-toside isoperistaltic techniques we can satisfy two goals: no blind loops are created and the total number of intestinal sutures is significantly reduced. The most surprising aspect of our series concerns the bowel segments treated by SP. As previously described by Ozuner and colleagues46 for standard SP techniques, macroscopic evidence of CD was completely absent in previous SP sites in the great majority of reoperated patients. This result, described also by Michelassi and col-

18

Sampietro et al

Nonconventional Strictureplasty in Crohn’s Disease

leagues19 and Poggioli and colleagues,20 in fewer patients followed for much shorter periods, represents very strong proof of the rationale for these techniques. Of the 201 SPs performed, recurrence was observed on a previous SP site, 2 Heineke-Mikulicz and 1 ileocolic side-toside isoperistaltic SP, only in 3 patients during 10 years followup (1.5%), despite the fact that NCSPs were performed on very extended diseased bowel segments (up to 60 cm in length in our experience). Evidence of disease regression is achievable after simple resolution of the complications, in spite of large amounts of macroscopically inflamed bowel left in situ, has emerged from two other recent studies from our department.47,48 In two series of patients treated by conservative operation we have documented at ultrasonographic followup the normalization of the bowel wall thickening, the most typical and constant finding in CD, with strong impact both on clinical and surgical recurrence and, at postoperative measurements, a trend toward normalization of antioxidant compounds levels and plasmatic oxidative stress caused by bowel inflammation. At time of surgical recurrence, we noticed changes in L-phenotype in two patients and in B-phenotype in four patients, opening a debate on the Vienna Classification. A shift of behavior category has been recently reported49 with the rate of changes in B-phenotype remaining stable over time, with approximately one-fourth of patients changing behavior every 5 years. On the contrary, the L-phenotype is considered to be more stable, but we recorded a change also in location. It should be noted that these patients worsen the global prognosis of our series because they do not relapse on intestinal segments previously operated. Similarly, patients with B2phenotype that shifts at recurrence to B3 account for the borderline p value of B2 patients at the Cox’s proportional hazard model, a trend that seems in contrast with others in literature. The B-phenotype is probably one of the most difficult characteristics to correlate with clinical and surgical recurrence, either because we are already starting to understand its behavior during the clinical history, or because of the influence of drug therapy that was widely changed during the last decade.50 Notwithstanding that, in our opinion, the Vienna Classification is still a valid and simple method to uniformly classify patients with CD, and allows comparisons of data from different series or in collaborative studies. In the surgical recurrence time-to-event estimates there are some cases of early recurrence within 1 year from operation. This

J Am Coll Surg

occurred in patients #22-A1L1B2-male (369 days), #24-A1L4B2-male (200 days), #40-A1L1B2-male (111 days), and #53-A1L1B2-male (356 days). Someone could object that in those patients there were no indications for an NCSP procedure. But similar cases are also present in all the resective and conservative series. In addition, patient #53 had a shift of CD location (L13L3), with an indication for operation because of a colonic localization. All of these patients had a previous wide intestinal resection and were ideal candidates for bowel-sparing techniques. We define clinical recurrence as the presence of abdominal symptoms that require medium-high doses of steroids. It is important to note that more than half of the patients did not require such a treatment during the 10 years. Considering the fact that the patient’s quality of life after surgery is now considered one of the primary concerns of surgical treatment,5,51-53 such data may eliminate many prejudices toward type of postoperative therapy in patients undergoing SPs and are even more satisfying than many results obtained in resective series. Neither the univariate analysis nor the Cox’s proportional hazard model have evidenced any prognostic factor able to influence clinical or surgical recurrence. This puts the inflammatory bowel disease surgeon in front of the impossibility of obtaining indications over those patients in whom it would be better not to perform a standard or nonconventional SP, but after all ad impossibilia nemo tenetur. In other words, these data seem to support our idea that at the moment of operation, with the exception of specific contraindications, there is no single factor that suggests which technique is best to use. To date, the existing conclusions about CD are that it is incurable and potentially panintestinal, that the majority of patients are likely to undergo one or more operations during their clinical history, that the risk for operation is ever-present and unpredictable in any given patient, that macroscopic disease at resection margins does not influence surgical recurrence,54,55 that no postoperative adjuvant therapy can prevent longterm recurrence, and that efforts to identify patients at certain risk of recrudescence have been unsuccessful for the heterogeneity of classification criteria and definitions used in the literature and for the intrinsic characteristics of CD. Unfortunately, whether the diseased, nonfunctioning bowel treated by SP turns into normal functioning bowel still remains the unanswered question about conservative operation in CD. The most popular position

Vol. 199, No. 1, July 2004

Sampietro et al

Nonconventional Strictureplasty in Crohn’s Disease

among gastrointestinal surgeons is probably that, if there were any evidence that nonfunctioning bowel treated by SP could be turned into functioning bowel, they would be convinced to perform SP as standard approach. In our opinion, and on the basis of studies in the literature, in the surgical approach toward CD the gastrointestinal surgeon should be guided by the opposite question: why should the diseased bowel be resected if there is no assurance that it could not turn into functioning bowel?

17.

Author Contributions

18.

Study conception and design: Sampietro, Cristaldi Acquisition of data: Sartani Analysis and interpretation of data: Sampietro Drafting of manuscript: Sampietro Critical revision: Cristaldi, Maconi, Parente, Danelli Statistical expertise: Sampietro Supervision: Porro, Taschieri Clinical followup: Maconi, Ardizzone

14. 15. 16.

19. 20. 21.

22. REFERENCES 1. Fazio RG, Whelan G, Fazio VW. Long-term follow-up of patients with Crohn’s disease. Gastroenterology 1985;88:1818– 1825. 2. Veloso FT, Ferreira JT, Barros L, Almeida S. Clinical outcome of Crohn’s disease: analysis according to the Vienna classification and clinical activity. Inflamm Bowel Dis 2001;7:306–313. 3. Sales DJ, Kirsner JB. Prognosis of inflammatory bowel disease. Arch Intern Med 1983;143:294–299. 4. Hellers G. Crohn’s disease in Stockholm County 1955–1974. A study of epidemiology, results of surgical treatments and long term prognosis. Acta Chir Scand 1979;490[Suppl 490]:31–69. 5. Krupnick AS, Morris JB. The long-term results of resection and multiple resections in Crohn’s disease. Semin Gastrointest Dis 2000;11:41–51. 6. Alexander-Williams J. The technique of intestinal strictureplasty. Int J Colorectal Dis 1986;1:54–57. 7. Fazio VW, Galandiuk S, Jagelman DG, Lavery IC. Strictureplasty in Crohn’s disease. Ann Surg 1989;210:621–625. 8. Dehn TC, Kettlewell MG, Mortensen NJ, et al. Ten-year experience of strictureplasty for obstructive Crohn’s disease. Br J Surg 1989;76:339–341. 9. Spencer MP, Nelson H, Wolff BG, Dozios RR. Strictureplasty for obstructive Crohn’s disease: the Mayo experience. Mayo Clin Proc 1994;69:33–36. 10. Serra J, Cohen Z, McLeod RS. Natural history of strictureplasty in Crohn’s disease: 9-year experience. Can J Surg 1995;38:481– 485. 11. Baba S, Nakai K. Strictureplasty for Crohn’s disease in Japan. J Gastroenterol 1995;30[Suppl 8]:135–138. 12. Hurst RD, Michelassi F. Strictureplasty for Crohn’s disease: techniques and long-term results. World J Surg 1998;22:359– 363. 13. Yamamoto T, Bain IM, Allan RN, Keighley MR. An audit of

23.

24. 25. 26.

27. 28. 29. 30. 31. 32.

33.

19

strictureplasty for small-bowel Crohn’s disease. Dis Colon Rectum 1999;42:797–803. Dietz DW, Laureti S, Strong SA, et al. Safety and longterm efficacy of strictureplasty in 314 patients with obstructing small bowel Crohn’s disease. J Am Coll Surg 2001;192:330–337. Pritchard TJ, Schoetz DJ Jr, Caushaj FP, et al. Strictureplasty of the small bowel in patients with Crohn’s disease. An effective surgical option. Arch Surg 1990;125:715–717. Tonelli F, Ficari F. Strictureplasty in Crohn’s disease: surgical option. Dis Colon Rectum 2000;43:920–926. Taschieri AM, Cristaldi M, Elli M, et al. Description of new bowel sparing techniques for long strictures of Crohn’s disease. Am J Surg 1997;173:509–512. Michelassi F. Side-to-side isoperistaltic strictureplasty for multiple Crohn’s strictures. Dis Colon Rectum 1996;39:345–349. Michelassi F, Hurst RD, Melis M, et al. Side-to-side isoperistaltic strictureplasty in extensive Crohn’s disease: a prospective longitudinal study. Ann Surg 2000;232:401–408. Poggioli G, Stocchi L, Laureti S, et al. Conservative surgical management of terminal ileitis. Dis Colon Rectum 1997;40: 234–239. Gasche C, Scholmerich J, Brynskov J, et al. A simple classification of Crohn’s disease: report of the Working Party for the World Congresses of Gastroenterology, Vienna 1998. Inflamm Bowel Dis 2000;6:8–15. Sampietro GM, Cristaldi M, Porretta T, et al. Early perioperative results and surgical recurrence after strictureplasty and miniresection for complicated Crohn’s disease. Dig Surg 2000;17: 261–267. Cristaldi M, Sampietro GM, Danelli PG, et al. Long-term results and multivariate analysis of prognostic factors in 138 consecutive patients operated on for Crohn’s disease using “bowelsparing” techniques. Am J Surg 2000;179:266–270. Delaney CP, Fazio VW. Crohn’s disease of the small bowel. Surg Clin North Am 2001;81:137–158, ix. Michelassi F, Balestracci T, Chappell R, Block GE. Primary and recurrent Crohn’s disease. Experience with 1379 patient. Ann Surg 1991;214:230–240. Romagnuolo J, Fedorak RN, Dias VC, et al. Hyperhomocysteinemia and inflammatory bowel disease: prevalence and predictors in a cross-sectional study. Am J Gastroenterol 2001;96: 2143–2149. Vasilopoulos S, Saiean K, Emmons J, et al. Terminal ileum resection is associated with higher plasma homocysteine levels in Crohn’s disease. J Clin Gastroenterol 2001;33:132–136. Thompson JS. Strategies for preserving intestinal length in the short bowel syndrome. Dis Colon Rectum 1987;30:208–213. Yamamoto T, Allan RN, Keighley MR. Long-term outcome of surgical management for diffuse jejunoileal Crohn’s disease. Surgery 2001;129:96–102. Dietz DW, Fazio VW, Laureti S, et al. Strictureplasty in diffuse Crohn’s jejunoileitis: safe and durable. Dis Colon Rectum 2002; 45:764–770. Buchman AL. Complications of long-term home total parenteral nutrition. Their identification, prevention and treatment. Dig Dis Sci 2001;46:1–18. Yamamoto T, Keighley MR. Factors affecting the incidence of postoperative septic complications and recurrence after strictureplasty for jejunoileal Crohn’s disease. Am J Surg 1999;178: 240–245. Gardiner KR, Kettlewell MG, Mortensen NJ. Intestinal haem-

20

34. 35. 36.

37. 38. 39. 40.

41. 42.

43. 44. 45. 46. 47.

48.

49. 50. 51.

52.

Sampietro et al

Nonconventional Strictureplasty in Crohn’s Disease

orrhage after strictureplasty for Crohn’s disease. Int J Colorectal Dis 1996;11:180–182. Ozuner G, Fazio VW. Management of gastrointestinal bleeding after strictureplasty for Crohn’s disease. Dis Colon Rectum 1995;38:297–300. Polito JM II, Childs B, Mellits ED, et al. Crohn’s disease: influence of age at diagnosis on site and clinical type of disease. Gastroenterology 1996;111:580–586. Post S, Herfarth C, Bohm E, et al. The impact of disease pattern, surgical management, and individual surgeons on the risk for relaparotomy for recurrent Crohn’s disease. Ann Surg 1996;223: 253–260. Yamamoto T, Allan RN, Keighley MR. Risk factors for intraabdominal sepsis after surgery in Crohn’s disease. Dis Colon Rectum 2000;43:1141–1145. Wolff BG. Factors determining recurrence following surgery for Crohn’s disease. World J Surg 1998;22:364–369. Greenstein AJ, Lachman P, Sachar DB, et al. Perforating and non-perforating indications for repeated operations in Crohn’s disease: evidence for two clinical forms. Gut 1988;29:588–592. Aeberhard P, Berchtold W, Riedtmann HJ, et al. Surgical recurrence of perforating and nonperforating Crohn’s disease. A study of 101 surgically treated patients. Dis Colon Rectum 1996;39: 80–87. Lautenbach E, Berlin JA, Lichtenstein GR. Risk factors for early postoperative recurrence of Crohn’s disease. Gastroenterology 1998;115:259–267. Poggioli G, Laureti S, Pierangeli F, Ugolini F. A new model of strictureplasty for multiple and long stenoses in Crohn’s ileitis: side-to-side diseased to disease-free anastomosis. Dis Colon Rectum 2003;46:127–130. Broering DC, Eisenberger CF, Koch A, et al. Strictureplasty for large bowel stenosis in Crohn’s disease: quality of life after surgical therapy. Int J Colorectal Dis 2001;16:81–87. Silverman RE, McLeod RS, Cohen Z. Strictureplasty in Crohn’s disease. Can J Surg 1989;32:19–22. Ozuner G, Fazio VW, Lavery IC, et al. How safe is strictureplasty in the management of Crohn’s disease? Am J Surg 1996; 171:57–60. Ozuner G, Fazio VW, Lavery IC, et al. Reoperative rates for Crohn’s disease following strictureplasty. Long-term analysis. Dis Colon Rectum 1996;39:1199–1203. Maconi G, Sampietro GM, Cristaldi M, et al. Preoperative characteristics and postoperative behavior of bowel wall on risk of recurrence after conservative surgery in Crohn’s disease: a prospective study. Ann Surg 2001;233:345–352. Sampietro GM, Cristaldi M, Cervato G, et al. Oxidative stress, vitamin A and vitamin E behaviour in patients submitted to conservative surgery for complicated Crohn’s disease. Digest Liv Dis 2002;34:696–701. Louis E, Collard A, Oger AF, et al. Behaviour of Crohn’s disease according to the Vienna classification: changing pattern over the course of the disease. Gut 2001;49:777–782. Lichtenstein GR. Treatment of fistulizing Crohn’s disease. Gastroenterology 2000;119:1132–1147. Thirlby RC, Land JC, Fenster LF, Lonborg R. Effect of surgery on health-related quality of life in patients with inflammatory bowel disease: a prospective study. Arch Surg 1998;133:826– 832. Thirlby RC, Sobrino MA, Randall JB. The long-term benefit of surgery on health-related quality of life in patients with inflammatory bowel disease. Arch Surg 2001;136:521–527.

J Am Coll Surg

53. Broering DC, Eisenberger CF, Koch A, et al. Quality of life after surgical therapy of small bowel stenosis in Crohn’s disease. Digest Surg 2001;18:124–130. 54. Raab Y, Bergstro¨ m R, Ejerblad S, et al. Factors influencing recurrence in Crohn’s disease: an analysis of a consecutive series of 353 patients treated with primary surgery. Dis Colon Rectum 1996;39:918–925. 55. Fazio VW, Marchetti F, Curch JM, et al. Effect of resections margins on the recurrences of Crohn’s Disease in the small bowel. A randomized controlled trial. Ann Surg 1996;224:563– 573.

Invited Commentary Fabrizio Michelassi, MD, FACS Chicago, IL Emmanuel Lee from Oxford introduced the concept of strictureplasty in the surgical treatment of Crohn’s disease in the early 1980s. This concept was then popularized by Alexander-Williams. Strictureplasty techniques avoid resection of diseased bowel and rely on the ability to successfully suture diseased bowel. Initially, there were concerns about high dehiscence and hemorrhagic rate and, considering that diseased bowel was left behind, there were concerns about increased recurrence rates. These concerns were eventually put to rest by large experiences showing a low perioperative morbidity and a recurrence rate that appears to be no higher than after a resection. With more experience gained, more advance bowelsparing techniques were described. The side-to-side isoperistaltic strictureplasty is one such technique, which we first performed in 1992 and first described in 1996. The concept is simple in that this strictureplasty is nothing but a long enteroenterostomy, but its application is technically challenging because the anastomosis is made with diseased bowel. The authors, like us and others earlier, have demonstrated that the side-to-side isoperistaltic strictureplasty is a safe technique and with a limited followup of slightly more than 3 years they have observed only a handful of recurrences. These are encouraging data that confirm existing data. I have three questions for the authors. The first question is technical: frequently the mesentery of the involved intestine is quite thickened and inflamed. What techniques do you use to transect it safely and does the thickness of the transected mesentery im-