Persistent perineal sinus after proctocolectomy for Crohn's disease

1 downloads 0 Views 604KB Size Report
The records of 145 patients who underwent proctocolectomy for Crohn's disease between 1970 and 1997 were reviewed. RESULTS: Persistent sinus occurred ...
Persistent Perineal Sinus After Proctocolectomy for Crohn's Disease Takayuki Yamamoto, M.D.,* Iain M. Bain, F.R.C.S.,* Robert N. Allan, F.R.C.P.,t Michael R. B. Keighley, F.R.C.S.* From the *University Department of Surgery and lhe tDepartment of Gastroenlerology, Queen Elizabeth Hospital, Edgbaston, Birmingham, United Kingdom PURPOSE: Persistent perineal sinus is a source of morbidity after proctocolectomy for Crotm's disease. This study examined the factors responsible for persistent sinus after proctocolectomy for Crolm's disease. We also assessed the outcome of surgical treatment for persistent pmineal sinus. METHODS: The records of 145 patients who underwent proctocolectomy for Crohn's disease between 1970 and 1997 were reviewed. RESULTS:Persistent sinus occurred in 33 (23 percent) patients after proctocolectomy. Factors associated with a significantly greater risk of perlneal sinus were younger age (P = 0.006), rectal involvement (P = 0.02), perianal sepsis (P = 0.0005), high fistulas (P = 0.04), extrasphincteric excision (P = 0.0004), and fecal contamination at operation (P = 0.0003). Multivariate analyses showed that age (P = 0.0001), rectal involvement (P = 0.007), and fecal contamination (P = 0.009) w e r e significant independent predictive factors for perineal sinus. Fifty-six operations, including 24 radical excisions, two rectus abdominis flaps, four gracilis transpositions, and two omentoplasties were performed in 24 patients with persistent sinus, but only 9 achieved healing. Long sinuses (> t0 cm) and sinuses presenting late (>12 weeks after proctocolectomy) were seldom cured by surgical treatment. CONCLUSION: Persistent perineal sinus is more likely to occur if an extrasphincteric dissection is needed because of extensive anorectal disease or if fecal contamination occurs at operation. Attempted surgical eradication of perineal sinus is often ineffective. [Key words: Crotm's disease; Perineal wound healing; Persistent perineal sinus; Proctocolectomy; Extrasphincteric dissection; intersphincteric dissection; Rectus abdominis flap; Gracilis transposition; Omentoplasty]

had a proctocolectomy, with particular reference to the m e t h o d of rectal excision. We also assessed the o u t c o m e o f surgical treatment for persistent perineaI sinus. PATIENTS

AND

METHODS

The records of 147 patients with Crohn's colitis treated b y p r o c t o c o l e c t o m y b e t w e e n J u n e 1, 1970, and D e c e m b e r 31, 1997, w e r e reviewed. O n e patient was lost to follow-up and another patient died within o n e m o n t h of p r o c t o c o l e c t o m y o f intra-abdominal sepsis; b o t h w e r e e x c l u d e d from this study. Perineal w o u n d healing was classified as follows: 1. early healing, within 12 w e e k s of proctocolectomy; 2. delayed healing, b e t w e e n 12 w e e k s and six m o n t h s of p r o c t o c o l e c t o m y ; a n d 3. persistent sinus, perineal w o u n d that remains u n h e a l e d for longer than six months. Comparisons b e t w e e n g r o u p s w e r e analyzed with the chi-squared test with Yates' correction. To determine i n d e p e n d e n t factors responsible for persistent sinus, multiple regression analyses w e r e performed. P < 0.05 was considered to be statistically significant.

Yamamoto T, Bain IM, Allan RaN, Keighley MRB. Persistent perineal sinus after proctocolectomy for Crohn's disease. Dis Colon Rectum 1999;42:96-101.

RESULTS A

discharging, persistent perineal w o u n d is a source of considerable morbidity af*er proctocolectomy a n d is m o r e c o m m o n after operation for Crohn's disease than for ulcerative colitis. 1-5 Several factors have b e e n associated with delayed healing or n o n h e a l i n g of perineal w o u n d s . These include gender, age, perioperative steroids, w o u n d m a n a g e m e n t , the degree of contamination at operation, and the presence of a high fistula-in-ano. ~' 6-10 W e reviewed o u r experience o f perineal w o u n d healing in patients with Crohn's disease w h o have

Mean age at p r o c t o c o l e c t o m y was 36 (range, 1482) years. There w e r e 48 males, Eighty-seven patients h a d a single-stage p r o c t o c o l e c t o m y , 24 patients und e r w e n t a c o l e c t o m y and ileorectal anastomosis but required a p r o c t e c t o m y later, and 34 patients h a d a c o l e c t o m y a n d retained rectal stump, w h i c h w a s rem o v e d later. Primary closure of the perineal w o u n d with suction drainage o f the pelvic space was p e r f o r m e d after p r o c t o c o l e c t o m y except w h e n there was contamination of the perineal w o u n d . The perineal w o u n d was left o p e n if fecal contamination o c c u r r e d during the operation. In 83 (57 percent) patients the perineal w o u n d

Read at the meeting of The American Society of Colon and Rectal Surgeons, San Antonio, Texas, May 2 to 7, 1998. No reprints are available. 96

Vol. 42, No. 1

PERINEALSINUS IN CROHN'S DISEASE

completely- healed within 12 weeks of proctocolectomy and in 29 (20 percent) the wound eventually healed, but healing was delayed. In 33 (23 percent) there was a persistent sinus. The median duration of follow-up after proctocolectomy was 157 (range, 1-275) months. Factors associated with persistent sinus were assessed (Table 1).

Age and Gender Patients were divided into three groups according to age at the time of proctocotectomy; younger age (-45 years) groups. Fourteen (35 percent) of 40 patients in the younger age group developed persistent sinus compared with 17 (26 percent) of 67 in the middle age group and 2 (6 percent) of 38 in the older age group (Table 1). Thus, persistent sinus was sig-

97

nificantly associated with younger age. However, there was no significant relationship between gender and perineal wound healing (Table 1).

Smoking Habit At the time of proctocolectomy, 61 patients smoked and 79 did not smoke (including 2 who formerly smoked). Smoking habits were unknown in five patients, who were excluded from this analysis. Smoking habits did not significantly affect perineal wound healing after surgery (Table 1).

Preoperative Corticosteroids Therapy Seventy-one (49 percent) patients received steroids for Crohn's disease before surgery. There was no significant difference in wound healing between pa-

Table 1.

Factors Associated with Persistent Sinus After Proctocolectomy for Crohn's Disease No Persistent Sinus Early Healing (n = 83)

Delayed Healing (n = 29)

Age at the time of operation (yr) -45 26 (68) 10 (26) Gender Male 24 (50) 10 (21) Female 59 (60) 19 (20) Smoking habit (-) 44 (55) 17 (22) (+) 37 (60) 12 (20) Unknown 2 0 Preoperative steroids therapy (-) 45 (61) 11 (15) (+) 38 (54) 18 (25) Rectal involvement (-) 20 (74) 6 (22) (+) 63 (53) 23 (20) Perianal sepsis (-) 61 (69) 16 (18) (+) 22 (39) 13 (22) High fistula-in-ano or rectovaginal fistula (-) 65 (61) 22 (21) (+) 18 (46) 7 (18) Method of rectal excision Intersphincteric dissection 49 (78) 9 (14) Extrasphincteric dissection 34 (42) 20 (24) Fecal contamination at the time of operation (-) 79 (68) 19 (t6) (+) 4 (14) 10 (36) Figures are number of patients and (percentage) unless otherwise specified. * No persistent sinus vs. persistent sinus groups.

Persistent Sinus (n = 33)

P Value*

14 (35) 17 (26) 2 (6)

0.006

14 (29) 19 (20)

0.28

18 (23) 12 (20) 3

0.81

18 (24) 15 (21)

0.79

1 (4) 32 (27)

0.02

11 (13) 22 (39)

0.0005

19 (18) 14 (36)

0.04

5 (8) 28 (34)

0.0004

19 (16) 14 (50)

0.0003

98

YAMA_MOTOE T A L

tients who had steroid therapy and those who did not (Table 1).

Rectal Involvement Rectal involvement, defined as the presence of pus, blood, ulceration, rigidity, loss of distensibility, or strictures by endoscopy or radiology or both, was present in 118 patients (81 percent). Thirty-two (97 percent) of the 33 patients with persistent perineal sinus had rectal involvement before proctocolectomy (Table 1). Persistent sinus occurred in 32 of 118 (27 percent) patients having rectal disease compared with 1 of 27 (4 percent) without rectal disease. There was a significant correlation between the presence of rectal involvement and perineal sinus.

Pefianal Sepsis Preoperative perianat sepsis was present in 57 (39 percent.) patients. Twenty-two (67 percent) of the 33 patients with persistent perineal sinus had perianal sepsis compared with 35 of 112 (31 percent) without persistent sinus (Table 1), Persistent sinus occurred in 22 of 57 (39 percent) patients with preoperative perianal sepsis compared with 11 of 88 (13 percent) without perianal sepsis. The presence of preoperative perianal sepsis significantly influenced the rate of persistent pefineal sinus.

High Fistula-in-Ano or Rectovaginal Fistula Preoperatively, 17 (12 percent) patients had a high fistula-in-ano, t9 (13 percent) had a rectovaginal fistula, and 3 (2 percent) had both. Fourteen (36 percent) of 39 patients having preoperative high fistulas (either in ano or rectovaginat) developed perineal sinus compared with 19 of 106 (18 percent) patients without high fistulas; this difference was significant (Table 1).

Methods of Rectal Excision Sixty-three (43 percent) patients underwent intersphincteric excision and 82 (57 percent) had an extrasphincteric excision. Extrasphincteric method was more frequently used on patients with anorectal disease such as perianal sepsis and high fistulas; 56 of 71 (79 percent) patients with anorectal disease compared with 26 of 74 (35 percent) without anorectal disease (P < 0.0001). Persistent perineal sinus occurred after 28 (34 percent) of 82 extrasphincteric excisions compared with only 5 (8 percent) of 63 intersphincteric

Dis Colon Rectum,January 1999

excisions (Table 1). In the patients who required an extrasphincteric proctectomy, persistent sinus occurred more frequently than in patients treated by intersphincteric excision.

Fecal Contamination at the T i m e o f Operation Fecal contamination caused by disease or accidental damage to the rectum during proctocolectomy occurred in 28 (19 percent) patients. After proctocolectomy 14 of 28 (50 percent) patients having fecal contamination developed a persistent perineal sinus compared with 19 of 117 (16 percent) patients without fecal contamination (Table 1). Fecal contamination significantly increased the risk of persistent perineal sinus.

Multivariate Analysis Multiple regression analyses showed that age at the time of operation (P = 0.0001), rectal involvement (P = 0.007), and fecal contamination at operation (P = 0.009) were significant independent predictive factors for perineal sinus. Other factors such as gender (P = 0.28), smoking habits (P = 0.31), steroids (P = 0.12), perianal sepsis (P = 0.09), high fistulas (P = 0.20), and method of rectal excision (P = 0.08) were not independent significant factors.

O u t c o m e of Surgical Treatment for P e r i n e a l Sinus Fifty-six surgical attempts had been made to cure a persistent sinus in 24 patients: drainage alone in 24, radical excision in 24 (including 2 coccygectomies), rectus abdominis flap in 2, gracilis transposition in 4, and omentoplasty in 2 cases. The sinus was resolved by radical excision and closure in 7 of 24 patients, by rectus abdominis flap in 1 of 2, and omentoplasty in 1 of 2. We compared attempted surgical treatment, length of sinus, and time at presentation of the sinus between the cured sinuses and uncured sinuses (Table 2). The surgical treatment was similar between the groups. Only 2 of 14 long sinuses (>10 cm) were resolved by surgical treatment compared with 7 of 10 short sinuses (-< 10 cm). Only 1 of 11 sinuses presenting late (>12 weeks after proctocolectomy) was resolved compared with 8 of 13 sinuses presenting earlier (----- 12 weeks after proctocolectomy).

PERINEAL SINUS IN CROHN'S DISEASE

Vol. 42, No. 1

99

Table 2. Comparisons of Cured and Uncured Sinuses after Surgical Treatment Cured Sinus (n = 9 )

Uncured Sinus (n = 15)

Attempted surgical treatment 9 15 Drainage alone 9* 15 Radical excision and primary closure 1" 1 Rectus abdominis flap 0 4 Gracilis transposition Omentoplasty 1" 1 Length of sinus ---10 cm 7 (70) 3 (30) >10 cm 2 (14) 12 (86) Time at presentation of sinus _12 weeks of proctocolectomy 1 (9) 10 (91) Figures are number of patients and (percentage) unless otherwise specified. * Seven radical excisions, one rectus abdominis flap, and one omentoplasty were successful.

DISCUSSION Persistent perineal sinus is much more common after proctocolectomy for Crohn's disease than in ulcerative colitis or malignant disease. 1-5 Corman e t al. 3 reported that six months after proctocolectomy, 72 percent of patients with Crohn's disease had unhealed perineal wounds compared with 56 percent of those with ulcerative colitis. In a previous article from our unit, 4 persistent perineal sinus occurred in 14 percent of patients with carcinoma, 31 percent of those with ulcerative colitis, and 36 percent of those with Crohn's disease. In this study we included only patients with Crohn's disease. Persistent sinus occurred in 23 percent of the patients. So far several factors have been reported to delay perineal w o u n d healing and thus predispose to persistent perineal sinus after proctocolectomy for inflammatory bowel disease. These factors include gender, younger age of patients, use of perioperative steroids, packing the wound with gauze, and fecal contamination at proctocolectomy. 1' 7-10 Jalan e t a L e concluded that primary perineal closure and suction drainage were the most significant factors in reducing the incidence of persistent perineal sinus. In a previous article from our unit, 6 healing was impaired in patients with a high fistulas (either in ano or rectovaginal) preoperatively, fecal contamination, and postoperative perineal sepsis. In the present study persistent sinus was associated with younger age (-< 25 years), rectal disease, perianal sepsis, high fistulas (either in ano or rectovaginal), fecal contamination at surgery, and extrasphincteric dissection. The gender,

P Value

0.62

0.02

0.03

smoking habit, or preoperative steroids had no adverse effect on wound healing. Multivariate analyses showed that age at the time of operation, rectal involvement, and fecal contamination were independent significant predictive factors for perineal sinus. Other factors such as perianal sepsis, high fistulas, and method of rectal excision were not independent significant factors, because there was a significant correlation among them. In 1972 Lee and Dowling 11 described perimuscular excision of the rectum for inflammatory bowel disease, and in 1977 Lyttle and Parks 12 described intersphincteric dissection of the rectum. Lubbers 5 reported that delayed w o u n d healing occurred at six months in only 4 of 27 patients undergoing intersphincteric resection. He stated that the intersphincteric method was effective, especially for patients with Crohn's disease. Leicester e t al. 13 reported perineal wound healing in 50 percent of the cases after 98 proctocolectomies with use of the intersphincteric method with healing at three months in 69 percent of cases. Berry e t al. 14 reported perineal wound healing in 75 percent of 115 patients after proctocolectomy by perimuscular dissection. However, they did not show how many patients developed a persistent sinus. Our current practice is to perform an intersphincteric dissection with primary suture of the wound, but if there is gross fecal contamination or established sepsis, the wound is left open. The extrasphincteric method is often necessary in patients with perianal sepsis or high fistulas. ~M:terintersphincteric dissection persistent perineal sinus occurred in 8 percent of 63

100

Y_&MAMOTO E T A L

cases compared with 34 percent of 82 cases after extrasphincteric dissection. A persistent perineal sinus is typically a long, fibrous tract covered by infected granulation tissue with a narrow external opening. Most patients will have troublesome symptoms of pain and discharge. Persistent symptomatic perineal sinus represents a challenge for surgical management. It is important to exclude a primary cause for delayed healing, such as a foreign body, an unrecognized pilonidal sinus, retained rectal mucosa, or an enteroperineal fistula. Simple curettage of the sinus with biopsy to allow adequate drainage is rarely successful alone. If granulomas are present, steroid treatment and even antituberculous therapy or immunosupression may occasionally achieve healing. ~5 If the sinus persists, radical excision of the sinus with closure of the w o u n d has been suggested. Ferrari and Den Besten 16 described seven patients treated by this method; healing was achieved within two weeks in all patients. In this study we performed this procedure in 29 patients (with partial coccygectomy in 2), but healing was only achieved in 7 patients. More extensive procedures for this condition include gracilis muscle transposition, rectus abdominis tram flap, and saucerization of the w o u n d with skin grafting. Ryan 17 reported that wide excision, including coccygectomy and transposition of the gracilis muscle flap, were successful in 12 of 15 patients (9 with ulcerative colitis and 6 with Crohn's disease). None of the patients treated by gracilis muscle transposition in this unit had a successful outcome. Rectus abdominis tram flap and omentoplasty were sometimes effective, but the number of patients was too small to form a conclusion about the efficacy of these procedures. The problem with the rectus abdominis flap is that it may compromise the strength of the abdominal incision and may compromise management of stoma complications or recurrent ileal disease, particularly if the stoma needs to be resited. By contrast, omentoplasty has none of these disadvantages, but both procedures involve an extensive abdominopefineal exposure. Long perineal sinuses ( > 10 cm) seldom could be eliminated (14 percent) despite any surgical treatment, whereas short sinuses were eliminated by reoperation in 70 percent of cases. Reconstructive surgery was unsuccessful in most patients (91 percent) w h o d e v e l o p e d purulent discharge from the perineal w o u n d s more than 12 weeks after proctocolectomy. These data question the role of aggres-

Dis Colon Rectum, January 1999

sive surgery for persistent perineal sinus in Crohn's disease, especially for long sinus and sinus presenting late. Patients requiring proctocolectomy for Crohn's disease must be counseled preoperatively about the risk of persistent perineal sinus, particularly if they have known risk factors such as youth, severe rectal disease, perianal sepsis, or complex fistulas requiring extrasphincteric excision where contamination is inevitable. The results of surgery fbr persistent perineal sinus after proctocolectomy were disappointing.

REFERENCES 1. Broader JH, Masselink BA, Oates GD, AlexanderWilliams J. Management of the pelvic space after proctectomy. BrJ Surg 1974;61:94-7. 2. Irvin TT, Goligher JC. A controlled clinical trial of three different methods of perineal wound management following excision of the rectum. Br J Surg 1975;62: 287-91. 3. Corman ML, Veidenheimer MC, Coller JA, Ross VH. Perineal wound healing after proctectomy for inflammatory bowel disease. Dis Colon Rectum 1978;21: 155-9. 4. Baudot P, Keighley MR, Alexander-Williams J. Perineal wound healing after proctectomy for carcinoma and inflammatory disease. Br J Surg 1980;67:275-6. 5. Lubbers E-JC. Healing of the perineaI wound after proctectomy for nonmalignant conditions. Dis Colon Rectum 1982;25:351-7. 6. Scammell BE, Keighley MR. Delayed perineal wound healing after proctectomy for Crohn's disease. BrJ Surg 1986;73:150-2. 7. Watts JM, de Dombal Fr, Goligher JC. Long-term complications and prognosis following major surgery for ulcerative colitis. BrJ Surg 1966;53:1014-23. 8. Jalan KN, Smith AN, Ruckley CV, Falconer CW, Small WP, Prescott RJ. Perineal wound healing in ulcerative colitis. Br J Surg 1969;56:749-53. 9. de DombaI FT, Burton I, GoligherJC. The early and late results of surgical treatment for Crohn's disease. Br J Surg 1971;58:805~40. 10. Marks CG, Ritchie JK, Todd IP, Wordsworth J. Primary suture of the perineal wound following rectal excision for inflarmnatory bowel disease. Br J Surg 1978;65: 560~i. 11. Lee EC, Dowling BL. Perimuscular excision of the rectum for Crohn's disease and ulcerative colitis: a conservation technique. Br J Surg 1972;59:29-32. 12. Lyttle JA, Parks AG. Intersphincteric excision of the rectum. Br J Surg 1977;64:413-6. 13. Leicester RJ, Ritchie JK, Wadsworth J, Thompson JP,

Vol. 42, No. 1

PERINEAL SINUS IN CROHN'S DISEASE

Hawley PR. Sexual function and perineal wound healing after intersphincteric excision of the rectum for inflammatory bowel disease. Dis Colon Rectum 1984; 27:244--8. 14. Berry AR, De Campos R, Lee EC. Perineal and pelvic morbidity" following perimuscutar excision of the rectum for inflammatory bowel disease. BrJ Surg 1986;73: 675-7.

101

15. Allan A, Keighley MR. Management of perianal Crohn's disease. World J Surg 1988;12:198-202. 16. Ferrari BT, Den Besten L. The prevention and treatment of the persistent perineal sinus. World J Surg 1980;4: 167-72. 17. Ryan JA. Gracilis muscle flap for the persistent perineal sinus of inflammatory- bowel disease. Am J Surg 1984; 148:64-70.