Abstract
Aim
Early detection of surgical site infection (SSI) allows for appropriate management after Crohn's disease (CD) surgery. The aim of this study was to evaluate the usefulness of bacterial culture of postoperative drainage fluid after CD surgery.
Methods
This study included 110 patients with CD who underwent surgery with bowel resection between January 2010 and March 2020 at Osaka University Hospital. Patients with only perianal surgery or incomplete records were excluded. Risk factors for SSI were evaluated in the context of clinical findings, including bacterial culture of postoperative drainage fluid, and bacterial species related to SSI were also examined.
Results
Of 110 patients, 18 (16.4%) developed SSI. Organ/space SSI developed in six, and a positive bacterial culture of drainage fluid (D‐Posi) was found in five (83.3%). Of 104 patients without organ/space SSI, 31 (29.8%) were D‐Posi (P = .027). Similarly, 68.8% with incisional SSI were D‐Posi, whereas 26.6% without incisional SSI were D‐Posi (P = .0021). Multivariate analysis revealed that D‐Posi was an independent risk factor in both organ/space and incisional SSI. Bacterial examination showed that Pseudomonas aeruginosa and Enterococcus faecalis were significantly detected in patients with SSI.
Conclusion
This study suggests the usefulness of postoperative drainage fluid bacterial culture for early diagnosis of SSI after CD surgery.
Keywords: bacterial culture, CD, drainage fluid, SSI
We demonstrate that bacterial culture of postoperative drainage fluid was useful to predict surgical site infection in the early phase after Crohn's disease surgery. This study suggests that bacterial culture allowed accurate diagnosis of SSI and prevention of exacerbations. Also, bacterial examination showed that Pseudomonas aeruginosa and Enterococcus faecalis were significantly detected in patients with SSI.
1. INTRODUCTION
Crohn's disease (CD) involves chronic intestinal inflammation and differences in gut bacterial species compared with healthy controls. 1 , 2 Despite improvements in medical treatments, 70%–90% of patients will undergo surgery in their lifetime. 3 More than half of them will require surgical intervention within 15 y of previous operations because of recurrence or new lesions occurrence. 4 , 5 , 6 Postoperative complications may not only delay resumption of medication but also trigger recurrence and complicate subsequent surgeries because of adhesions. 7 , 8 If complications are detected in the early phase, exacerbations could be avoided and prognosis improved. 7 , 8 For these reasons, accurate predictors of complications are needed.
Surgical site infection (SSI) is one of the postoperative complications, and the rate of SSI after CD surgery is higher than after colorectal cancer surgery. 4 , 9 , 10 , 11 Several SSI risk factors have been identified in patients with CD, including anemia, a longer duration of surgery, and higher intraoperative lactate level. 4 , 11 Although these are indirect risk factors for SSI, a direct factor is bacterial contamination of the surgical site. 12 Few reports, however, have described the association between positive bacterial culture of lavage or drainage fluid and SSI in gastrointestinal surgery. 12 , 13 , 14 To our knowledge, no studies have evaluated the role of positive bacterial culture of drainage fluid (D‐Posi) after CD surgery.
We hypothesized that D‐Posi could predict SSI in the early phase after CD surgery. The aims of this study were to evaluate the clinical impact of D‐Posi in detecting SSI and the characteristics of detected bacterial species after CD surgery.
2. METHODS
2.1. Patients
A total of 245 consecutive patients who underwent CD surgery between January 2010 and March 2020 at Osaka University Hospital were included in this study. We excluded 125 patients who underwent perianal surgery only and 10 patients with incomplete records (Figure 1). Informed consent was obtained from all patients before the surgery. The study protocol was approved by the Institutional Review Board of Osaka University Hospital (# 15028).
FIGURE 1.
Flow chart of participant inclusion in the study
2.2. Perioperative management and surgical procedure
Prophylactic antibiotics with second‐generation cephalosporin (cefmetazole) were intravenously administered within 30 min before skin incision, repeatedly at 3‐h intervals during surgery, and postoperatively twice a day until postoperative day (POD) 2. Drainage tubes were placed in a Morrison's pouch or Douglas’ pouch. Before March 2012, we performed closed passive drainage using surgeon's choice of a SILASCON duple drain (Kaneka Medical Products, Japan) or closed active drainage using a BLAKE silicone drain (Ethicon, Cincinnati, OH). After April 2012, closed active drainage was used in all cases.
A wound protector was used during surgery, and intraperitoneal lavage was performed before wound closure. All surgical staff changed gloves after bowel anastomosis. The peritoneum was closed by a running suture with 3‐0 Vicryl (Ethicon). The muscular fascia was closed by a knotted suture with 1‐PDS PLUS (Ethicon). The wound was washed with 200 mL of saline before closure of the skin by a buried suture with 4‐0 PDS plus (Ethicon). Board‐certified surgeons performed all surgeries.
2.3. Outcome measurement and definition of SSI
For evaluating candidate risk factors for SSI, we collected medical records data on patient characteristics including age, sex, body mass index (BMI), American Society of Anesthesiologists score, duration of disease, behavior (nonstricturing and nonpenetrating, stricturing, or penetrating), preoperative laboratory data, antitumor necrosis factor alpha (TNF)‐α use, steroid use, and frequency of surgery. We also collected data on surgical variables, including type of surgery (open or laparoscopic), surgical site (small bowel, colon, ileocolon, or rectum), with or without abdominoperineal resection (APR), ostomy creation, surgical wound classification, duration of surgery, intraoperative bleeding, transfusion, and D‐Posi. The surgical staff examined the wounds daily, and bacterial culture of drainage fluid was performed on POD 1 and 4 in all cases. The definition of D‐Posi was detection of bacteria in the drainage fluid at POD 1 or 4.
Surgical site infection was defined according to the US Centers for Disease Control and Prevention classifications, 4 , 15 as an infection that occurred at the surgical site within 30 d after surgery with at least one of the following observations: purulent discharge from the incision or from the drain placed into the organ/space; bacteria isolated by culturing fluid or tissue from the incision or the organ/space; an open wound with signs and symptoms of infection; and/or an abscess or other evidence of infection found on examination of the incision or the organ/space. 11 , 16
2.4. Detection method of bacterial species
Bacterial culture of drainage fluid was performed on the same day a sample was submitted. The sample was smeared for detecting the type by gram staining, and then coated on both Agar’s and Ringer’s medium. Results of the gram staining were available the same day. The bacterial species were examined with matrix‐assisted laser desorption/ionization‐time‐of‐flight mass spectrometry using colonies on Agar medium. On the day following sample submission, D‐Posi status and bacterial species were identified. The antibiotic sensitivity was identified at 2 d after sample submission. If the amount of bacteria was limited, this step sometimes took a few more days and up to 1–2 wk for very low counts associated with using Ringer’s medium.
2.5. Statistical analysis
All statistical analyses were performed using JMP 15 (SAS Institute, Cary, NC).
The reference values were determined according to the median values about age, duration of disease, C‐reactive protein, duration of surgery, and intraoperative bleeding, and also according to the standard values of white blood cell (WBC), hemoglobin, platelet and albumin, and BMI. Patients’ backgrounds, surgical variables, and postoperative results were compared using the Chi‐squared or Fisher's exact test for categorical variables, as appropriate. Odds ratios (ORs) were used to evaluate the relationship between SSI and different factors. Variables from comparisons yielding P < .1 in the univariate analysis were entered into a multivariate analysis. For multivariate analysis, we used a logistic regression model to investigate the factors associated with SSI incidence. The predictive accuracy and area under the curve (AUC) were calculated. Receiver operating characteristic analyses were performed to compare the performance of variables as SSI risk predictors. Risk factors, excluding D‐Posi, with a univariate P < .1, were defined as “Others” and “Others” together with D‐Posi was defined as “Combination.” Diagnostic accuracy was judged using the diagnostic OR (DOR), which is not affected by disease frequency and incidence of complications. Statistical significance was defined as P < .05. The ratios were calculated as follows:
3. RESULTS
3.1. Patient characteristics
Data for 110 patients were analyzed and the characteristics of the 84 men and 26 women are shown in Table 1. The median age at surgery was 39.5 y, and 36 patients (32.7%) were D‐Posi. The APR accounted for 10 patients (9.1%). The median duration of surgery was 203.5 min, and the median intraoperative bleeding was 145 mL. Seven patients (6.4%) received a blood transfusion. A total of 17 patients (15.5%) received preoperative antibiotics for high inflammatory response with penetrating disease.
TABLE 1.
Patient characteristics (N = 110)
| N* | |
|---|---|
| Preoperative | Median (range) |
| Age, y | 39.5 (14–72) |
| Male/female | 84/26 |
| BMI, kg/m2 | 19.0 (14.0–28.4) |
| ASA score ≥2, n (%) | 71 (64.5%) |
| Duration of disease, y | 10 (0–41) |
| Behavior | |
| Nonstricturing & nonpenetrating | 7 (6.4%) |
| Stricturing | 41 (37.2%) |
| Penetrating | 62 (56.4%) |
| Preoperative WBC, per µL | 5260 (1990–16 890) |
| Preoperative Hb, g/dL | 12.0 (7.3–16.1) |
| Preoperative platelet ×104, per µL | 28.8 (6.8–57.8) |
| Preoperative albumin, g/dL | 3.5 (1.5–4.9) |
| Preoperative CRP, mg/dL | 0.20 (0–22.4) |
| Preoperative use of anti‐TNF‐α antibody | 78 (70.9%) |
| Preoperative use of steroid | 11 (10%) |
| Frequency of surgery (primary/redo) | 66/44 |
| Surgical | |
| Open/laparoscopic | 13/97 |
| Surgical site | |
| Small bowel/colon/ileocolon/rectum | 21/14/65/10 |
| APR, n (%) | 10 (9.1%) |
| Ostomy creation | 18 |
| Wound class (II/III/IV) | 95/8/7 |
| Duration of surgery, min | 203.5 (105–748) |
| Intraoperative bleeding, mL | 145 (0–8050) |
| Transfusion, n (%) | 7 (6.4%) |
| D‐Posi, n (%) | 36 (32.7%) |
Abbreviations: APR, abdominoperineal resection; ASA, American Society of Anesthesiologists; BMI, body mass index; CRP, C‐reactive protein; D‐Posi, bacterial culture positive of postoperative drainage fluid; Hb, hemoglobin; WBC, white blood cell count.
Median (range) unless otherwise specified.
3.2. Postoperative complications
Postoperative complications were diagnosed in 28 patients (25.5%), including overall SSI in 18 (16.4%), incisional SSI in 16 (14.5%), and organ/space SSI in six (5.5%) (12 had incisional SSI only, two had organ/space SSI only, and four had both types of SSI). Gastrointestinal bleeding occurred in five patients (4.5%), intra‐abdominal bleeding in one (0.9%), subcutaneous bleeding in one (0.9%), ileus in five (4.5%), catheter infection in four (3.6%), and femoral nerve palsy in one (0.9%). Severe complications (Clavien–Dindo grade ≥ III) were observed in seven patients (6.4%). There were no perioperative deaths.
3.3. SSI risk factors
For analyses, 18 patients were in the overall SSI group and compared with the other 92 patients in the nonoverall SSI group. Univariate analysis of risk factors associated with overall SSI identified preoperative WBC ≥9400/µL, open surgery, APR, a longer duration of surgery, a high degree of intraoperative bleeding, and D‐Posi (Table 2). Risk factors for incisional SSI were WBC ≥9400/µL, APR, a longer duration of surgery, and D‐Posi; the only risk factor associated with organ/space SSI was D‐Posi. Of the patients with incisional SSI, 68.8% (11/16) were D‐Posi, whereas 26.6% (25/ 94) without incisional SSI were D‐Posi (P = .0021). Similarly, 83.3% (5/6) with organ/space SSI were D‐Posi, and 29.8% (31/104) without organ/space were D‐Posi (P = .027) (Table 3). In multivariate analyses, only D‐Posi emerged as an independent risk factor of overall, incisional, and organ/space SSI. In addition, high BMI was also an independent risk factor for incisional SSI (Table 4).
TABLE 2.
Univariate analysis of risk factors associated with overall surgical site infection (SSI)
|
Overall SSI group, n (%) (n = 18) |
Nonoverall SSI group, n (%) (n = 92) |
OR | 95% CI | P | |
|---|---|---|---|---|---|
| Preoperative backgrounds | |||||
| Age ≥40 y | 10 (55.6%) | 45 (48.9%) | 1.31 | 0.47–3.60 | .607 |
| Male | 15 (83.3%) | 69 (75.0%) | 1.67 | 0.44–6.28 | .450 |
| BMI ≥22 kg/m2 | 7 (38.9%) | 21 (22.8%) | 2.15 | 0.74–6.24 | .159 |
| ASA score ≥2 | 12 (66.7%) | 59 (64.1%) | 1.12 | 0.38–3.26 | .837 |
| Duration of disease ≥10 y | 9 (50.0%) | 47 (51.6%) | 0.94 | 0.34–2.57 | .898 |
| Behavior | |||||
| Penetrating/nonpenetrating | 8/10 | 54/38 | 0.56 | 0.20–1.56 | .269 |
|
Preoperative WBC ≥9.4 × 103/µL |
5 (27.8%) | 9 (9.8%) | 3.55 | 1.03–12.3 | .045 a |
|
Preoperative Hb <12 g/dL |
12 (66.7%) | 43 (46.7%) | 2.28 | 0.79–6.59 | .129 |
|
Preoperative platelet ≥32 × 104/µL |
9 (50.0%) | 31 (33.7%) | 1.97 | 0.71–5.46 | .193 |
|
Preoperative albumin <3.5 g/dL |
9 (52.9%) | 38 (41.8%) | 1.57 | 0.55–4.44 | .396 |
|
Preoperative CRP ≥0.2 mg/dL |
11 (61.1%) | 43 (46.7%) | 1.79 | 0.64–5.03 | .269 |
| Preoperative use of anti‐TNF‐α antibody | 13 (72.2%) | 65 (70.7%) | 1.08 | 0.35–3.33 | .893 |
| Preoperative use of steroid | 2 (11.1%) | 9 (9.8%) | 1.15 | 0.22–5.84 | .864 |
| Frequency of surgery; redo | 9 (50.0%) | 35 (38.0%) | 1.63 | 0.59–4.50 | .347 |
| Surgical backgrounds | |||||
| Open/Laparoscopic | 5/13 | 8/84 | 4.04 | 1.14–14.2 | .030 a |
| Surgical site | |||||
| Small bowel | 3 | 18 | 0.25 | 0.04–1.45 | .123 |
| Colon | 2 | 12 | 0.25 | 0.04–1.77 | .166 |
| Ileocolon | 9 | 56 | 0.24 | 0.06–1.03 | .054 |
| Rectum | 4 | 6 | Reference = 1 | ||
| APR | 4 (22.2%) | 6 (6.52%) | 4.10 | 1.02–16.4 | .046 a |
| Ostomy creation | 5 (27.8%) | 13 (14.1%) | 2.34 | 0.71–7.66 | .161 |
| Wound class ≥III | 3 (16.7%) | 12 (13.0%) | 1.33 | 0.34–5.30 | .683 |
|
Duration of surgery ≥204 min |
14 (77.8%) | 41 (44.6%) | 4.35 | 1.33–14.2 | .015 a |
|
Intraoperative bleeding ≥145 mL |
13 (72.2%) | 42 (45.7%) | 3.10 | 1.02–9.39 | .046 a |
| Transfusion | 2 (11.1%) | 5 (5.4%) | 2.18 | 0.39–12.2 | .377 |
| D‐Posi | 13 (72.2%) | 23 (25.0%) | 7.80 | 2.51–24.2 | .0004 a |
Abbreviations: 95% CI, 95% confidence interval; APR, abdominoperineal resection; ASA, American Society of Anesthesiologists; BMI, body mass index; CRP, C‐reactive protein; D‐Posi, bacterial culture–positive postoperative drainage fluid; Hb, hemoglobin; OR, odds ratio; WBC, white blood cell count.
P < .05
TABLE 3.
Univariate analysis of risk factors associated with incisional surgical site infection (SSI) and organ space SSI
| Incisional SSI | Organ/space SSI | |||||||
|---|---|---|---|---|---|---|---|---|
|
Incisional SSI, n (%) (n = 16) |
Nonincisional SSI, n (%) (n = 94) |
OR (95% CI) |
P |
Organ/space SSI, n (%) (n = 6) |
Nonorgan/space SSI, n (%) (n = 104) |
OR (95% CI) |
P | |
| Preoperative backgrounds | ||||||||
| Age ≥40 y |
10 (62.5%) |
45 (47.9%) |
1.8 (0.61–5.40) |
.284 |
1 (16.7%) |
54 (51.9%) |
0.19 (0.02–1.64) |
.130 |
| Male |
14 (87.5%) |
70 (74.5%) |
2.4 (0.51–11.34) |
.269 |
4 (66.7%) |
80 (76.9%) |
0.60 (0.10–3.48) |
.569 |
|
BMI ≥22 kg/m2 |
7 (43.8%) |
21 (22.3%) |
2.7 (0.90–8.13) |
.077 |
1 (16.7%) |
27 (26.0%) |
0.57 (0.06–5.10) |
.616 |
|
ASA score ≥2 |
10 (62.5%) |
61 (64.9%) |
0.90 (0.30–2.70) |
.853 |
4 (66.7%) |
67 (64.4%) |
1.1 (0.19–6.32) |
.911 |
|
Duration of disease ≥10 y |
8 (50.0%) |
48 (51.6%) |
0.94 (0.32–2.71) |
.905 |
4 (66.7%) |
52 (50.5%) |
2.0 (0.34–11.18) |
.448 |
| Behavior | ||||||||
|
Penetrating/ nonpenetrating |
6/10 | 56/38 |
0.41 (0.14–1.21) |
.11 | 4/2 | 58/48 |
1.59 (0.28–9.05) |
.604 |
|
Preoperative WBC ≥9.4 × 103/µL |
5 (31.3%) |
9 (9.6%) |
4.3 (1.22–15.15) |
.024 a |
1 (16.7%) |
13 (12.5%) |
1.4 (0.15–12.95) |
.767 |
|
Preoperative Hb <12 g/dL |
10 (62.5%) |
45 (47.9%) |
1.8 (0.61–5.40) |
.284 |
4 (66.7%) |
51 (49.0%) |
2.1 (0.36–11.85) |
0.410 |
|
Preoperative platelet ≥32 × 104/µL |
9 (56.3%) |
31 (33.0%) |
2.6 (0.89–7.67) |
.081 |
2 (33.3%) |
38 (36.5%) |
0.87 (0.15–4.97) |
.874 |
|
Preoperative albumin <3.5 g/dL |
7 (46.7%) |
40 (43.0%) |
1.2 (0.39–3.46) |
.791 |
3 (50.0%) |
44 (43.1%) |
1.3 (0.25–6.85) |
.742 |
|
Preoperative CRP ≥0.2 mg/dL |
11 (68.8%) |
43 (45.7%) |
2.6 (0.84–8.10) |
.097 |
2 (33.3%) |
52 (50.0%) |
0.50 (0.09–2.85) |
.435 |
| Preoperative use of anti‐TNF‐α antibody |
11 (68.8%) |
67 (71.3%) |
0.89 (0.28–2.79) |
.837 |
6 (100%) |
72 (69.2%) |
NA | NA |
| Preoperative use of steroid |
2 (12.5%) |
9 (9.6%) |
1.35 (0.26–6.91) |
.719 |
1 (16.7%) |
10 (9.6%) |
1.90 (0.20–17.73) |
.581 |
| Frequency of surgery; redo |
8 (50.0%) |
36 (38.3%) |
1.61 (0.56–4.67) |
.380 |
3 (50.0%) |
41 (39.4%) |
1.54 (0.30–7.98) |
.609 |
| Surgical backgrounds | ||||||||
| Open/laparoscopic | 4/12 | 9/85 |
3.1 (0.84–11.83) |
.090 | 2/4 | 11/93 |
4.2 (0.69–25.8) |
.118 |
| Surgical site | ||||||||
|
Small bowel/ Colon or ileocolon/ rectum |
2 10 4 |
19 69 6 |
0.16 (0.02–1.09) 0.22 (0.05–0.91) Reference=1 |
.061 .036 a |
2 2 2 |
19 77 8 |
0.42(0.05–3.53) 0.10(0.01–0.84) Reference=1 |
.425 .034 a |
| APR |
4 (25.0%) |
6 (6.4%) |
4.9 (1.20–19.86) |
.027 a |
2 (33.3%) |
8 (7.7%) |
6.0 (0.95–37.94) |
.057 |
| Ostomy creation |
4 (25.0%) |
14 (25.9%) |
1.9 (0.53–6.76) |
.319 |
2 (33.3%) |
16 (15.4%) |
2.8 (0.46–16.29) |
.265 |
| Wound class ≥III |
2 (12.5%) |
13 (13.8%) |
0.89 (0.18–4.38) |
.886 |
1 (16.7%) |
14 (13.5%) |
1.29 (0.140–11.83) |
.824 |
|
Duration of surgery ≥204 min |
12 (75.0%) |
43 (45.7%) |
3.6 (1.07–11.84) |
.039 a |
5 (83.3%) |
50 (48.1%) |
5.4 (0.61–47.83) |
.130 |
|
Intraoperative bleeding ≥145 mL |
11 (68.8%) |
44 (46.8%) |
2.5 (0.81–7.76) |
.113 |
5 (83.3%) |
50 (48.1%) |
5.4 (0.61–47.83) |
.130 |
| Transfusion |
2 (12.5%) |
5 (5.3%) |
2.5 (0.45–14.4) |
.292 |
0 (0%) |
7 (6.7%) |
NA | NA |
| D‐Posi |
11 (68.8%) |
25 (26.6%) |
6.1 (1.92–19.21) |
.0021 a |
5 (83.3%) |
31 (29.8%) |
11.8 (1.32–105.0) |
.027 a |
Abbreviations: 95% CI, 95% confidence interval; APR, abdominoperineal resection; ASA, American Society of Anesthesiologists; BMI, body mass index; CRP, C‐reactive protein; D‐Posi, bacterial culture–positive postoperative drainage fluid; Hb, hemoglobin; NA, not applicable; OR, odds ratio; WBC, white blood cell count.
P < .05.
TABLE 4.
Multivariate analysis of risk factors associated with surgical site infection (SSI)
| OR | 95% CI | P | |
|---|---|---|---|
| Overall SSI | |||
| Preoperative WBC ≥9.4 × 103/µL | 1.83 | 0.40–8.44 | .439 |
| Open | 1.59 | 0.37–6.92 | .534 |
| APR | 1.82 | 0.32–10.3 | .495 |
| Duration of surgery ≥204 min | 2.49 | 0.55–11.3 | .237 |
| Intraoperative bleeding ≥145 mL | 1.30 | 0.31–5.48 | .722 |
| D‐Posi | 6.00 | 1.79–19.9 | .0036 a |
| Incisional SSI | |||
| BMI ≥22 kg/m2 | 4.77 | 1.17–19.5 | .030 a |
| Preoperative WBC ≥9.4 × 103/µL | 1.60 | 0.28–9.07 | .598 |
| Preoperative platelet ≥32 × 104/µL | 2.90 | 0.71–11.9 | .139 |
| Preoperative CRP ≥0.2 mg/dL | 1.52 | 0.32–7.23 | .598 |
| Open | 2.04 | 0.38–11.1 | .408 |
| APR | 2.91 | 0.43–19.9 | .276 |
| Duration of surgery ≥204 min | 1.78 | 0.39–8.13 | .451 |
| D‐Posi | 5.59 | 1.42–22.0 | .014 a |
| Organ/space SSI | |||
| APR | 4.63 | 0.65–33.0 | .126 |
| D‐Posi | 10.6 | 1.26–96.1 | .037 a |
Abbreviations: 95% CI, 95% confidence interval; APR, abdominoperineal resection; D‐Posi, bacterial culture–positive postoperative drainage fluid; OR, odds ratio; WBC, whiteblood cell count.
P < .05.
Abdominoperineal resection has been reported as a risk factor for incisional and organ/space SSI, leading to assess also the clinical impact of D‐Posi with exclusion of APR. 11 Of 100 patients with APR excluded, overall SSI developed in 14 patients: 10 with incisional SSI only, two with organ/space only, and two with both. Eleven (78.6%) with overall SSI were D‐Posi, whereas 20 (23.3%) without overall SSI were D‐Posi (P = .0004). We found that 75% with incisional SSI were D‐Posi and 22.0% without incisional SSI were D‐Posi (P = .002). Similarly, 75% with organ/space SSI were D‐Posi, and 29.2% without organ/space SSI were D‐Posi (P = .091). Multivariate analysis identified D‐Posi as an independent risk factor in overall SSI and incisional SSI also among the patients with APR excluded.
3.4. Treatment details for the case of D‐Posi
Of 110 patients, 36 were D‐Posi, 19 (52.8%) of whom received only prophylactic antibiotics. Four (11.1%) had longer administration of cefmetazole for a few days. Nine (25.0%) were switched to other antibiotics with broad‐spectrum activity, based on the detected bacterial species. Four (11.1%) were switched to other antibiotics and had percutaneous drainages. Among the patients with D‐Posi, confirming their status took a median of 2 d in 13 patients with SSI and 5 d in 23 patients with non‐SSI. Identifying antibiotic sensitivity took a median of 4 (2–17) d. The period for detecting D‐Posi was significantly shorter in the SSI group compared to the non‐SSI group (P = .007), possibly because of different bacterial loads.
3.5. Comparison of predictive and diagnostic accuracy
Next, we evaluated the predictive and diagnostic ability of D‐Posi. In overall SSI, the AUC for D‐Posi was 0.736, and the DOR was 7.81, and the AUC was higher than the “Others” group (AUC = 0.733). Values for the “Combination” were the highest, with an AUC of 0.819 and DOR of 21.3 (Figure 2A and Table 5). In organ/space SSI, the AUC for D‐Posi was 0.768, and the DOR was 11.8 (Figure 2C and Table 5), while D‐Posi was not a strong predictor for incisional SSI compared to “Others” and “Combination,” with an AUC of 0.711 and DOR of 6.07 (Figure 2B and Table 5). The negative predictive value was above 90% in all groups.
FIGURE 2.
Receiver operating characteristic curve analysis based on risk factors other than D‐Posi, D‐Posi, or a combination of both for surgical site infection (SSI). (A) Overall SSI. (B) Incisional SSI. (C) Organ/space SSI. AUC, area under the curve. D‐Posi, positive bacterial culture of drainage fluid
TABLE 5.
Diagnostic accuracies for predicting surgical site infection (SSI)
|
Sensitivity (%) |
Specificity (%) |
PPV (%) |
NPV (%) |
LR (+) | LR (−) | DOR | |
|---|---|---|---|---|---|---|---|
| Overall SSI | |||||||
| Others | 50.0 | 89.1 | 47.4 | 90.1 | 4.60 | 0.56 | 8.20 |
| D‐Posi | 72.2 | 75.0 | 36.1 | 93.2 | 2.89 | 0.37 | 7.81 |
| Combination | 77.8 | 85.9 | 51.9 | 95.2 | 5.50 | 0.26 | 21.3 |
| Incisional SSI | |||||||
| Others | 87.5 | 68.1 | 31.8 | 97.0 | 2.74 | 0.18 | 14.9 |
| D‐Posi | 68.8 | 73.4 | 30.6 | 93.2 | 2.59 | 0.43 | 6.07 |
| Combination | 81.3 | 86.2 | 50.0 | 96.4 | 5.88 | 0.22 | 27.0 |
| Organ/space SSI | |||||||
| Others | 33.3 | 92.3 | 20.0 | 96.0 | 4.33 | 0.72 | 6.00 |
| D‐Posi | 83.3 | 70.2 | 13.9 | 98.6 | 2.80 | 0.24 | 11.8 |
| Combination | 83.3 | 70.2 | 13.9 | 98.6 | 2.80 | 0.24 | 11.8 |
Abbreviations: DOR, diagnostic odds ratio; D‐Posi, bacterial culture–positive postoperative drainage fluid; LR (−), negative likelihood ratio; LR (+), positive likelihood ratio; NPV, negative predictive value; PPV, positive predictive value.
P < .05.
3.6. Characteristics of detected bacteria
We investigated which bacteria were associated with SSI risk. In univariate analysis, Pseudomonas aeruginosa and Enterococcus faecalis were significantly associated with overall SSI (P = .0024 and .017, respectively). Pseudomonas aeruginosa was frequently detected in patients with incisional and organ/space SSI Klebsiella pneumoniae and E. faecalis were often frequently detected in patients with organ/space SSI (Table S1). The comparison of detected bacteria between SSI and D‐Posi were also performed. Of 16 patients with incisional SSI, seven (43.8%) had the same organism detected from both samples. Among six patients with organ/space SSI, four (66.6%) had the same organism in both samples (Table S2). Of 18 patients with SSI, six organisms were subjected to an antibiotic sensitivity test of both samples. Most of the antibiotic sensitivity results were the same in both samples, and the examined organisms were resistant to the prophylactic antibiotic cefmetazole (Table S3).
Of 110 patients, only four patients underwent bacterial culture of intraoperative samples (Proteus vulgaris detected in one case, Bacteroides fragilis in one, K. pneumoniae and Escherichia coli in one, and none detected in the fourth case). Of these four patients, two were D‐Posi, but the detected bacteria were different from the intraoperative samples, and none of the four patients developed SSI.
4. DISCUSSION
In patients with CD, it is important to detect SSI in the early phase and prevent worsening, because infectious complications could increase the recurrence risk. 7 , 8 , 17 , 18 Thus, accurate predictors of SSI are needed in this patient population. Several studies have suggested that D‐Posi could have predictive associations with predicting SSI. 12 , 14 Sugiura et al reported that bacterial culture‐positive peritoneal lavage fluid correlated with SSI and pancreatic fistula after pancreaticoduodenectomy. 12 Migita et al showed that D‐Posi was an independent risk factor among clinical and surgical variables for intra‐abdominal abscess after gastrectomy. 14 Routine use of prophylactic drainage has been reported not to decrease postoperative complications in colorectal surgeries, including ileocecal resection or right hemicolectomy. 19 , 20 However, undrained fluid could cause intra‐abdominal abscess because of bacterial leakage from the mesentery in patients with CD. 21 Compared with patients with cancer, patients with CD also have a higher risk of infection because of low nutrition, the use of immunosuppressive drugs, and the presence of fistulas and abscess. Therefore, we routinely placed prophylactic drainage for our patients with CD. This study is the first to investigate the impact of D‐Posi and detected bacteria in these patients. The results identified D‐Posi as an independent predictor of SSI, with an AUC >0.8 in the early phase.
Organ/space SSI usually becomes evident around POD 7 or later. 22 , 23 To prevent worsening, early administration of antibiotics is recommended for suspected cases after gastrointestinal surgery. 24 , 25 Percutaneous drainage also can preclude the need for surgical intervention for intra‐abdominal abscess. 26 In our study, six patients had organ/space SSI, and all were diagnosed at POD 8 (POD 5–15) because computed tomography diagnosis of organ/space SSI was mainly performed following symptom progression. The negative predictive value of diagnosing SSI was >95% with the use of D‐Posi, suggesting its potential to allow prediction of SSI on POD 4. Of six patients with organ/space SSI, four with D‐Posi were switched to other antibiotics with broad‐spectrum activity on POD 4.5 (POD 2–7), and one patient continued receiving antibiotics from 14 d before surgery until POD 11 because of high inflammation with penetrating disease. Identifying the D‐Posi status might allow us perform empirical treatment in the early phase before the definitive diagnosis.
Several studies have demonstrated that patients with CD have gut bacterial alterations characterized by reduced bacterial diversity, depletion of Firmicutes or Bacteroides, and enrichment of Proteobacteria. 1 , 2 , 27 , 28 , 29 However, the causative bacteria of SSI in patients with CD have not been reported. A previous study showed that E. faecalis, S. aureus, and P. aeruginosa were often isolated from SSI following lower intestinal tract surgery. 30 Our data also showed that E. faecalis and P. aeruginosa were significantly detected in CD patients with organ/space SSI. Because P.aeruginosa affected immunosuppressed patients, careful management would be essential for CD patients under treatment. 31 , 32 The current findings suggest that organisms at the site of SSI were resistant to the prophylactic antibiotic cefmetazole and that most of the antibiotic sensitivity results were the same between drainage fluid and of SSI bacterial culture. This suggested that sensitivity results in drainage fluid could guide the choice of an antibiotic.
The present study had several limitations. First, these results were based on a single‐center retrospective cohort study, including some selection bias that could not be avoided. Second, the impact of preoperative medicines, such as biologics except anti‐TNF‐α therapy, immune suppressants, and 5‐aminosalicylic acid were not evaluated in this study. A larger study that includes these factors is needed.
In conclusion, this study confirmed the clinical usefulness of D‐Posi for SSI and demonstrated the association of bacteria species with CD surgery. Bacterial culture of drainage fluid represents an easy, noninvasive, and inexpensive tool to perform appropriate management of SSI.
DISCLOSURES
The study protocol was approved by the Institutional Review Board of Osaka University Hospital (# 15028).
Informed consent was obtained from all patients before the surgery.
This research was not preregistered in an independent, institutional registry (N/A).
This research was not an animal study (N/A).
The authors T.O. and T.M. were supported by the KINSHUKAI GROUP.
Author T.M. was partly supported by a research grant from the Osaka Medical Research Foundation for Intractable Disease, JSP KAKENHI (20K17615), Danone Institute of Japan Foundation (DIJF R01‐24), and Astellas Pharma Global Development (Grant No. IDRS2017A0000577).
AUTHOR CONTRIBUTIONS
Study concept and design: M.I., T.O., and T.M. Acquisition of data: M.I., T.O., and T.M. Statistical analysis and interpretation of data: M.I., T.O., and M.F. Drafting of the article: M.I. and T.O. Critical revision of the article for important intellectual content: all authors. Supervision: Y.D. and H.E.
Supporting information
Table S1‐S3
ACKNOWLEDGMENT
The authors thank the staff at Osaka University Hospital.
Ichihara M, Ogino T, Fujii M, Haraguchi N, Takahashi H, Miyoshi N, et al. Usefulness of Bacterial Culture of Drainage Fluid for Predicting Surgical Site Infection After Crohn’s Disease Surgery. Ann Gastroenterol Surg. 2022;6:375–385. doi: 10.1002/ags3.12530
Funding information
This study was partly supported by a research grant from the Osaka Medical Research Foundation for Intractable Disease, JSPS KAKENHI (20K17615), Danone Institute of Japan Foundation (DIJF R01‐24), and Astellas Pharma Global Development (Grant No. IDRS2017A000577)
REFERENCES
- 1. Machiels K, Pozuelo Del Río M, Martinez‐De la Torre A, Xie Z, Pascal Andreu V, Sabino J, et al. Early postoperative endoscopic recurrence in Crohn's disease is characterised by distinct microbiota recolonisation. J Crohns Colitis. 2020;14(11):1535–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Wright EK, Kamm MA, Wagner J, Teo SM, Cruz P, Hamilton AL, et al. Microbial factors associated with postoperative Crohn's disease recurrence. J Crohns Colitis. 2017;11(2):191–203. [DOI] [PubMed] [Google Scholar]
- 3. Byrne J, Stephens R, Isaacson A, Yu H, Burke C. Image‐guided percutaneous drainage for treatment of post‐surgical anastomotic leak in patients with Crohn's disease. J Crohns Colitis. 2016;10(1):38–42. [DOI] [PubMed] [Google Scholar]
- 4. Liu S, Miao J, Wang G, Wang M, Wu X, Guo K, et al. Risk factors for postoperative surgical site infections in patients with Crohn's disease receiving definitive bowel resection. Sci Rep. 2017;7(1):9828. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Olaisen M, Flatberg A, Granlund AVB, Røyset ES, Martinsen TC, Sandvik AK, et al. Bacterial mucosa‐associated microbiome in inflamed and proximal noninflamed ileum of patients with Crohn's disease. Inflamm Bowel Dis. 2021;27(1):12–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Ogino T, Mizushima T, Matsuda C, Mori M, Doki Y. Essential updates 2018/2019: Colorectal (benign): Recent updates (2018–2019) in the surgical treatment of benign colorectal diseases. Ann Gastroenterol Surg. 2020;4(1):30–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Riss S, Schuster I, Papay P, Mittlböck M, Stift A. Repeat intestinal resections increase the risk of recurrence of Crohn's disease. Dis Colon Rectum. 2013;56:881–7. [DOI] [PubMed] [Google Scholar]
- 8. Guo Z, Cao L, Guo F, Gong J, Li Y, Gu L, et al. The presence of postoperative infectious complications is associated with the risk of early postoperative clinical recurrence of Crohn's disease. World J Surg. 2017;41(9):2371–7. [DOI] [PubMed] [Google Scholar]
- 9. Bhakta A, Tafen M, Glotzer O, Ata A, Chismark AD, Valerian BT, et al. Increased incidence of surgical site infection in IBD patients. Dis Colon Rectum. 2016;59(4):316–22. [DOI] [PubMed] [Google Scholar]
- 10. Uchino M, Ikeuchi H, Tsuchida T, Nakajima K, Tomita N, Takesue Y. Surgical site infection following surgery for inflammatory bowel disease in patients with clean‐contaminated wounds. World J Surg. 2009;33(5):1042–8. [DOI] [PubMed] [Google Scholar]
- 11. Uchino M, Ikeuchi H, Bando T, Chohno T, Sasaki H, Horio Y, et al. Efficacy of preoperative oral antibiotic prophylaxis for the prevention of surgical site infections in patients with Crohn disease: a randomized controlled trial. Ann Surg. 2019;269(3):420–6. [DOI] [PubMed] [Google Scholar]
- 12. Sugiura T, Mizuno T, Okamura Y, Ito T, Yamamoto Y, Kawamura I, et al. Impact of bacterial contamination of the abdominal cavity during pancreaticoduodenectomy on surgical‐site infection. Br J Surg. 2015;102(12):1561–6. [DOI] [PubMed] [Google Scholar]
- 13. Minervini S, Bentley S, Youngs D, Alexander‐Williams J, Burdon DW, Keighley MR. Prophylactic saline peritoneal lavage in elective colorectal operations. Dis Colon Rectum. 1980;23(6):392–4. [DOI] [PubMed] [Google Scholar]
- 14. Migita K, Takayama T, Matsumoto S, Wakatsuki K, Tanaka T, Ito M, et al. Impact of bacterial culture positivity of the drainage fluid during the early postoperative period on the development of intra‐abdominal abscesses after gastrectomy. Surg Today. 2014;44(11):2138–45. [DOI] [PubMed] [Google Scholar]
- 15. Allegranzi B, Zayed B, Bischoff P, Kubilay NZ, de Jonge S, de Vries F, et al. New WHO recommendations on intraoperative and postoperative measures for surgical site infection prevention: an evidence‐based global perspective. Lancet Infect Dis. 2016;16(12):e288–303. [DOI] [PubMed] [Google Scholar]
- 16. Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitions for nosocomial infections, 1988. Am J Infect Control. 1988;16(3):128–40. [DOI] [PubMed] [Google Scholar]
- 17. Mizushima T, Nakajima K, Takeyama H, Naito A, Osawa H, Uemura M, et al. Single‐incision laparoscopic surgery for stricturing and penetrating Crohn's disease. Surg Today. 2016;46(2):203–8. [DOI] [PubMed] [Google Scholar]
- 18. Ikeshima R, Mizushima T, Takahashi H, Haraguchi N, Nishimura J, Hata T, et al. The efficacy of active drainage for preventing postoperative organ/space surgical site infections in patients with Crohn's disease. Surg Today. 2018;48(1):25–32. [DOI] [PubMed] [Google Scholar]
- 19. Zhang HY, Zhao CL, Xie J, Ye YW, Sun JF, Ding ZH, et al. To drain or not to drain in colorectal anastomosis: a meta‐analysis. Int J Colorectal Dis. 2016;31(5):951–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Yeh CY, Changchien CR, Wang JY, Chen JS, Chen HH, Chiang JM, et al. Pelvic drainage and other risk factors for leakage after elective anterior resection in rectal cancer patients: a prospective study of 978 patients. Ann Surg. 2005;241(1):9–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Ha CWY, Martin A, Sepich‐Poore GD, Shi B, Wang Y, Gouin K, et al. Translocation of viable gut microbiota to mesenteric adipose drives formation of creeping fat in humans. Cell. 2020;183(3):666–83.e17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Tanaka K, Kumamoto T, Nojiri K, Takeda K, Endo I. The effectiveness and appropriate management of abdominal drains in patients undergoing elective liver resection: a retrospective analysis and prospective case series. Surg Today. 2013;43(4):372–80. [DOI] [PubMed] [Google Scholar]
- 23. Torzilli G, Olivari N, Del Fabbro D, Gambetti A, Leoni P, Gendarini A, et al. Bilirubin level fluctuation in drain discharge after hepatectomies justifies long‐term drain maintenance. Hepatogastroenterology. 2005;52(64):1206–10. [PubMed] [Google Scholar]
- 24. Kamiya S, Hiki N, Kumagai K, Honda M, Nunobe S, Ohashi M, et al. Two‐point measurement of amylase in drainage fluid predicts severe postoperative pancreatic fistula after gastric cancer surgery. Gastric Cancer. 2018;21(5):871–8. [DOI] [PubMed] [Google Scholar]
- 25. Fritz S, Hartwig W, Lehmann R, Will‐Schweiger K, Kommerell M, Hackert T, et al. Prophylactic antibiotic treatment is superior to therapy on‐demand in experimental necrotising pancreatitis. Crit Care. 2008;12(6):R141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Clancy C, Boland T, Deasy J, McNamara D, Burke JP. A meta‐analysis of percutaneous drainage versus surgery as the initial treatment of Crohn's disease‐related intra‐abdominal abscess. J Crohns Colitis. 2016;10(2):202–8. [DOI] [PubMed] [Google Scholar]
- 27. Liguori G, Lamas B, Richard ML, Brandi G, da Costa G, Hoffmann TW, et al. Fungal dysbiosis in mucosa‐associated microbiota of Crohn's disease patients. J Crohns Colitis. 2016;10(3):296–305. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Manichanh C, Borruel N, Casellas F, Guarner F. The gut microbiota in IBD. Nat Rev Gastroenterol Hepatol. 2012;9(10):599–608. [DOI] [PubMed] [Google Scholar]
- 29. Khor B, Gardet A, Xavier RJ. Genetics and pathogenesis of inflammatory bowel disease. Nature. 2011;474(7351):307–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Takesue Y, Kusachi S, Mikamo H, Sato J, Watanabe A, Kiyota H, et al. Antimicrobial susceptibility of pathogens isolated from surgical site infections in Japan: Comparison of data from nationwide surveillance studies conducted in 2010 and 2014–2015. J Infect Chemother. 2017;23(6):339–48. [DOI] [PubMed] [Google Scholar]
- 31. Sadikot RT, Blackwell TS, Christman JW, Prince AS. Pathogen‐host interactions in Pseudomonas aeruginosa pneumonia. Am J Respir Crit Care Med. 2005;171(11):1209–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Matsunaga T, Miyata H, Sugimura K, Asukai K, Yanagimoto Y, Takahashi Y, et al. Clinical usefulness of a perioperative bacteriological culture to treat patients with postoperative pneumonia after esophagectomy. Ann Gastroenterol Surg. 2018;3(1):57–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1‐S3