Journal of Crohn's and Colitis, 2015, 259–265 doi:10.1093/ecco-jcc/jju029 Original Article
Original Article
The Impact of Bacterial DNA Translocation on Early Postoperative Outcomes in Crohn’s Patients Undergoing Abdominal Surgery Yi Li, Lugen Zuo, Weiming Zhu, Jianfeng Gong, Wei Zhang, Lili Gu, Zhen Guo, Ning Li, Jieshou Li Department of General Surgery, Jinling Hospital, Medical School of Nanjing University, No. 305 East Zhongshan Road, Nanjing, PR China Corresponding author: Weiming Zhu, MD, Department of General Surgery, Jinling Hospital, Medical School of Nanjing University, No. 305 East Zhongshan Road, Nanjing 210002, China.
Abstract Background and Aims: Bacterial DNA (bactDNA) translocation occurs frequently in patients with Crohn’s disease and can be present in patients with a negative blood microbiological culture. We aimed to determine the effects of bactDNA translocation on postoperative outcomes in Crohn’s disease patients undergoing abdominal surgery. Methods: Patients with Crohn’s disease who underwent abdominal surgery between January 2012 and March 2014 were identified. General and postoperative outcome-related information was retrieved from a database, and the data were compared between patients with and without bactDNA translocation. Multivariate analysis was used to determine the independent effect of bactDNA translocation on postoperative morbidity. Results: One hundred and seven patients who underwent abdominal surgery were included in our study. The presence of bactDNA in blood samples was identified in 29 patients (27.1%). There was a total of 55 complications in 28 patients (26.2%). Patients with bactDNA in their blood had a mean postoperative hospital stay of 12.7 ± 4.2 days and patients without DNA translocation had a mean postoperative hospital stay of 10.1 ± 4.8 days (p = 0.009). The readmission rate was increased in patients with bactDNA translocation (p = 0.032). A low preoperative level of serum albumin (p = 0.024), preoperative immunosuppressive agent use (p = 0.046), and the presence of bactDNA in blood (p = 0.005) were independently associated with increased postoperative adverse outcomes. Conclusions: Preoperative bactDNA translocation into the blood increases the incidence of postoperative adverse outcomes in patients with Crohn’s disease who undergo abdominal surgery. Keywords: Crohn’s disease; bacterial DNA; postoperative complication; abdominal surgery
1. Introduction Crohn’s disease is a relapsing and chronic inflammatory disorder of the gastrointestinal tract. It is thought to result from aberrant and overly aggressive immune responses to commensal gut microorganisms in the genetically susceptible host.1,2 Damage to the intestinal barrier function leading to bacterial translocation is one important proposed
mechanism of intestinal inflammation in Crohn’s disease. The baseline rate of translocation in humans was reported to be 5–10%.3 However, the incidence of cultured bacteria in mesenteric lymphatic nodes was higher in Crohn’s disease patients than in control subjects.4–6 This result implies there is a higher rate of bacterial translocation in patients with Crohn’s disease. In addition, intestinal bacterial translocation to the mesenteric adipose tissue is increased in patients with Crohn’s disease.7
Copyright © 2015 European Crohn’s and Colitis Organisation (ECCO). Published by Oxford University Press. All rights reserved. For permissions, please email: [email protected]
259
Y. Li et al.
260 Despite recent improvements in medical therapy, the majority of Crohn’s disease patients require surgical intervention to treat disease complications at some point during their disease course.8 Crohn’s disease patients frequently require abdominal surgery and may present more postoperative complications. Thus, it is important to identify the risk factors associated with postoperative complications and reduce postoperative morbidity to improve patient quality of life. Additionally, understanding risk factors may provide evidence for the timing of surgical intervention. Studies addressing the risk factors of postoperative outcomes have been conducted. The factors associated with postoperative complications in Crohn’s disease patients undergoing abdominal surgery include anti-tumor necrosis factor (TNF) therapy,9 immunosuppressive therapy,10,11 emergency surgery, and blood loss.12 However, studies of the impact of perioperative factors such as disease behavior, age, and immunomodulation on postoperative outcomes in Crohn’s disease patients have yielded conflicting results.13–15 Crohn’s disease patients have a greater incidence of translocation of bacteria into lymphatic nodes than non-Crohn’s disease controls. This increase may result in a high rate of postoperative septic complications in Crohn’s disease patients.4,16 The presence of bacterial DNA (bactDNA) in blood is recognized as a marker of bacterial translocation.17 BactDNA translocation was found in a subgroup of inflammatory bowel disease (IBD) patients with negative blood bacterial cultures,18 and the presence of bactDNA was associated with an increased inflammatory response.19,20 Bacterial translocation also increases the development of postoperative sepsis in patients receiving surgical intervention.16,21 BactDNA translocation in Crohn’s disease patients may result in a systemic inflammatory response, especially in patients who have undergone surgery and have a subsequent high rate of postoperative morbidity. However, the effect of bactDNA translocation on postoperative outcomes in Crohn’s disease patients remains unknown. Therefore, the aim of our study was to investigate the impact of preoperative bactDNA translocation on postoperative outcomes in patients receiving abdominal surgery for Crohn’s disease. Our study analyzed a cohort of patients with Crohn’s disease and compared postoperative outcomes following abdominal surgery in patients with and without bactDNA in their blood.
2. Methods 2.1. Patients and samples Patients with Crohn’s disease who underwent abdominal surgery at the Inflammatory Bowel Disease Center of Jinling Hospital between January 2012 and March 2014 were reviewed and the data were gathered. Patient information was collected from a prospectively maintained IBD database. Only patients admitted to our institute for the purpose of surgical intervention for Crohn’s disease were included in this study. The detailed exclusion criteria were as follows: (1) presence of abdominal abscess/fistula or perianal abscess; (2) surgeries for perianal disease only; and (3) concomitant infections. The presence of an abdominal abscess/fistula or perianal abscess was documented preoperatively by radiological assessment (such as computed tomography and magnetic resonance imaging) and physical examination and was confirmed by intraoperative findings. Subjects were also excluded from the study if their blood samples had been mishandled or if they received antibiotic treatment in the preceding 2 weeks. Blood samples were obtained from patients 1 day before surgical intervention. As previously reported,18–20,22 blood samples were collected under aseptic conditions that prevented exposure to air. All patients included in our study provided informed consent. The study protocol was approved by the Ethics Committee of Jinling Hospital.
2.2. Standardized collection of patient records A standardized chart was used to retrieve patient data. The data were collected by two independent individuals who were blinded to the blood bactDNA results. The following data were collected: age, gender, body mass index (BMI), duration of disease, previous Crohn’s disease-related surgery, C-reactive protein (CRP) level, white blood count, serum albumin, disease location, preoperative drug use (immunosuppressive drugs), Crohn’s Disease Activity Index (CDAI) before surgery,23 indication for surgery, status of surgery (elective or emergency), laparoscopic surgery, smoking status, number of anastomoses, American Society of Anesthesiologists (ASA) wound classification, operative time, postoperative glucose value, and information regarding postoperative outcome. The blood glucose measurements were performed 48 h postoperatively and hyperglycemia was defined as glucose values more than 125 mg/dL.24
2.3. Detection of bactDNA fragments in blood BactDNA in blood samples was identified by conducting a broadrange PCR and then sequencing a conserved region of the 16S ribosomal RNA (16SrRNA) gene, as previously described.18–20,22 Blood sample DNA was extracted using the QIAamp DNA Mini Kit (Qiagen, 51304) according to the manufacturer’s instructions. The yield and purity of DNA were measured by reading A260 and A280 values in a Shimadzu spectrophotometer. The isolated DNA was used as a template to amplify the hypervariable V3 regions of the 16SrRNA gene. The universal PCR primers were V3F (5′-CCAGACTCCTACGGGAGGCAG-3′) and V3R (5′-CGTATTACCGCGGCTGCTG-3′). Positive and negative controls were evaluated in duplicate in each assay to avoid false-positive results. Genomic DNA extracted from the microbiota in ileum contents was added as a positive control. The negative control consisted of nuclease-free water (ABI, AM9930) and PCR mixture (without template). An aliquot of DNA (100 ng) was added to each reaction mixture. The PCR reactions were performed with a thermocycling program consisting of 35 cycles in a GeneAmp 9700 (Applied Biosystems, Foster City, CA) using the following protocol: 95°C for 40 s, 55°C for 40 s, and 72°C for 60 s. The resulting PCR amplicons were checked for the correct size (203 bp) on 2% agarose gels that were stained with ethidium bromide and visualized under ultraviolet light.
2.4. Postoperative outcomes We were particularly interested in the following 30-day postoperative outcomes: postoperative complications, time to bowel movement, postoperative blood transfusion requirement, duration of postoperative hospitalization, postoperative stay in the intensive care unit, readmission, and reoperation. Special attention was also paid to the organ dysfunction and systemic inflammatory response syndrome (SIRS).25 All complications were defined as those occurring within 30 days from the date of surgery. Early postoperative complications were divided into infectious and noninfectious complications. The infectious complications included wound infection, intra-abdominal abscess, anastomotic leak, pneumonia, urinary tract infection, and bacteremia. The noninfectious complications included postoperative ileus, small bowel obstructions, dehydration, and others.
2.5. Statistical analysis Continuous variables are presented as the mean ± SD and were compared using Student’s t-test for normally distributed variables. Fisher’s exact test or the χ2 test was used to compare categorical variables. The relative risks and 95% confidence intervals (CIs)
BactDNA and postoperative outcome in CD were also calculated to identify potential effects. Bivariate regression analysis was performed to determine the association between the presence of bactDNA and postoperative complications and to identify potential confounding variables associated with bactDNA translocation and postoperative outcomes. All covariates associated with postoperative morbidity were selected for possible inclusion in the corresponding multivariable models if the p-value for the bivariate association was less than 0.05. Multivariable models were then manually constructed in a forward stepwise manner. A p-value of less than 0.05 was considered statistically significant.
261 in blood versus 12.8% (10/78) in patients without bactDNA translocation (p = 0.023). There were wound infections in 8 of 29 (27.6%) patients who had bactDNA translocation compared with 7 of 78 (9.0%) patients who did not have bactDNA translocation (p = 0.025). There were no differences in the rates of intraabdominal abscess (p = 0.296), anastomotic leak (p = 0.617), pneumonia (p = 0.178), urinary tract infection (p = 0.060), or bacteremia (p = 0.470) between patients with and without bactDNA translocation.
3.4. Postoperative noninfectious complications
3. Results 3.1. Demographic and clinical characteristics at baseline A total of 107 patients with Crohn’s disease met the inclusion criteria and were included in this study. Twenty-nine patients (27.1%) showed the presence of bactDNA in their blood samples. BactDNA was not present in the blood of the other 78 patients (72.9%). In this study, none of the included patients presented a positive blood microbiological culture at inclusion. The mean age at surgery was 33.1 ± 11.7 years and the mean Crohn’s disease duration was 7.9 ± 4.7 years. Seventeen patients (15.9%) required a diverting stoma and there were 81 (75.7%) patients with one anastomosis. There were nine (8.4%) patients with more than one anastomosis. There were 18 (16.8%) patients who received emergency surgical intervention. The most common indication for surgery was stricturing/obstruction (82.2%), followed by medically refractory disease (17.8%). Before surgery, steroids were given in 42 (39.3%) cases and immunomodulators were used in 52 (48.6%) patients. There were 15 (14%) patients treated with anti-TNF agents. There was a higher mean level of preoperative CRP in patients with bactDNA in their blood samples (p = 0.001). There were no significant differences between the groups for the following variables: age, gender, duration of disease, previous Crohn’s disease-related abdominal surgery, BMI, laparoscopic surgery, preoperative concomitant treatment, indication for surgery, smoking status, urgent surgery, number of anastomoses, white blood count, albumin level, ASA score, wound classification, operative time, postoperative hyperglycemia, active disease with a CDAI of 150 or more, or Montreal classification (Table 1).
3.2. Early postoperative outcomes There were no postoperative deaths. There were 55 complications in 28 (26.2%) patients. Table 2 indicates that 28 complications were noted in 15 patients with bactDNA in their blood compared with 27 complications in 13 patients without bactDNA translocation (p 1 White blood count, ×109/L (mean ± SD) C-reactive protein, mg/L (mean ± SD) Albumin, g/L (mean ± SD) ASA score, n (%) ≥3
Original Article
The Impact of Bacterial DNA Translocation on Early Postoperative Outcomes in Crohn’s Patients Undergoing Abdominal Surgery Yi Li, Lugen Zuo, Weiming Zhu, Jianfeng Gong, Wei Zhang, Lili Gu, Zhen Guo, Ning Li, Jieshou Li Department of General Surgery, Jinling Hospital, Medical School of Nanjing University, No. 305 East Zhongshan Road, Nanjing, PR China Corresponding author: Weiming Zhu, MD, Department of General Surgery, Jinling Hospital, Medical School of Nanjing University, No. 305 East Zhongshan Road, Nanjing 210002, China.
Abstract Background and Aims: Bacterial DNA (bactDNA) translocation occurs frequently in patients with Crohn’s disease and can be present in patients with a negative blood microbiological culture. We aimed to determine the effects of bactDNA translocation on postoperative outcomes in Crohn’s disease patients undergoing abdominal surgery. Methods: Patients with Crohn’s disease who underwent abdominal surgery between January 2012 and March 2014 were identified. General and postoperative outcome-related information was retrieved from a database, and the data were compared between patients with and without bactDNA translocation. Multivariate analysis was used to determine the independent effect of bactDNA translocation on postoperative morbidity. Results: One hundred and seven patients who underwent abdominal surgery were included in our study. The presence of bactDNA in blood samples was identified in 29 patients (27.1%). There was a total of 55 complications in 28 patients (26.2%). Patients with bactDNA in their blood had a mean postoperative hospital stay of 12.7 ± 4.2 days and patients without DNA translocation had a mean postoperative hospital stay of 10.1 ± 4.8 days (p = 0.009). The readmission rate was increased in patients with bactDNA translocation (p = 0.032). A low preoperative level of serum albumin (p = 0.024), preoperative immunosuppressive agent use (p = 0.046), and the presence of bactDNA in blood (p = 0.005) were independently associated with increased postoperative adverse outcomes. Conclusions: Preoperative bactDNA translocation into the blood increases the incidence of postoperative adverse outcomes in patients with Crohn’s disease who undergo abdominal surgery. Keywords: Crohn’s disease; bacterial DNA; postoperative complication; abdominal surgery
1. Introduction Crohn’s disease is a relapsing and chronic inflammatory disorder of the gastrointestinal tract. It is thought to result from aberrant and overly aggressive immune responses to commensal gut microorganisms in the genetically susceptible host.1,2 Damage to the intestinal barrier function leading to bacterial translocation is one important proposed
mechanism of intestinal inflammation in Crohn’s disease. The baseline rate of translocation in humans was reported to be 5–10%.3 However, the incidence of cultured bacteria in mesenteric lymphatic nodes was higher in Crohn’s disease patients than in control subjects.4–6 This result implies there is a higher rate of bacterial translocation in patients with Crohn’s disease. In addition, intestinal bacterial translocation to the mesenteric adipose tissue is increased in patients with Crohn’s disease.7
Copyright © 2015 European Crohn’s and Colitis Organisation (ECCO). Published by Oxford University Press. All rights reserved. For permissions, please email: [email protected]
259
Y. Li et al.
260 Despite recent improvements in medical therapy, the majority of Crohn’s disease patients require surgical intervention to treat disease complications at some point during their disease course.8 Crohn’s disease patients frequently require abdominal surgery and may present more postoperative complications. Thus, it is important to identify the risk factors associated with postoperative complications and reduce postoperative morbidity to improve patient quality of life. Additionally, understanding risk factors may provide evidence for the timing of surgical intervention. Studies addressing the risk factors of postoperative outcomes have been conducted. The factors associated with postoperative complications in Crohn’s disease patients undergoing abdominal surgery include anti-tumor necrosis factor (TNF) therapy,9 immunosuppressive therapy,10,11 emergency surgery, and blood loss.12 However, studies of the impact of perioperative factors such as disease behavior, age, and immunomodulation on postoperative outcomes in Crohn’s disease patients have yielded conflicting results.13–15 Crohn’s disease patients have a greater incidence of translocation of bacteria into lymphatic nodes than non-Crohn’s disease controls. This increase may result in a high rate of postoperative septic complications in Crohn’s disease patients.4,16 The presence of bacterial DNA (bactDNA) in blood is recognized as a marker of bacterial translocation.17 BactDNA translocation was found in a subgroup of inflammatory bowel disease (IBD) patients with negative blood bacterial cultures,18 and the presence of bactDNA was associated with an increased inflammatory response.19,20 Bacterial translocation also increases the development of postoperative sepsis in patients receiving surgical intervention.16,21 BactDNA translocation in Crohn’s disease patients may result in a systemic inflammatory response, especially in patients who have undergone surgery and have a subsequent high rate of postoperative morbidity. However, the effect of bactDNA translocation on postoperative outcomes in Crohn’s disease patients remains unknown. Therefore, the aim of our study was to investigate the impact of preoperative bactDNA translocation on postoperative outcomes in patients receiving abdominal surgery for Crohn’s disease. Our study analyzed a cohort of patients with Crohn’s disease and compared postoperative outcomes following abdominal surgery in patients with and without bactDNA in their blood.
2. Methods 2.1. Patients and samples Patients with Crohn’s disease who underwent abdominal surgery at the Inflammatory Bowel Disease Center of Jinling Hospital between January 2012 and March 2014 were reviewed and the data were gathered. Patient information was collected from a prospectively maintained IBD database. Only patients admitted to our institute for the purpose of surgical intervention for Crohn’s disease were included in this study. The detailed exclusion criteria were as follows: (1) presence of abdominal abscess/fistula or perianal abscess; (2) surgeries for perianal disease only; and (3) concomitant infections. The presence of an abdominal abscess/fistula or perianal abscess was documented preoperatively by radiological assessment (such as computed tomography and magnetic resonance imaging) and physical examination and was confirmed by intraoperative findings. Subjects were also excluded from the study if their blood samples had been mishandled or if they received antibiotic treatment in the preceding 2 weeks. Blood samples were obtained from patients 1 day before surgical intervention. As previously reported,18–20,22 blood samples were collected under aseptic conditions that prevented exposure to air. All patients included in our study provided informed consent. The study protocol was approved by the Ethics Committee of Jinling Hospital.
2.2. Standardized collection of patient records A standardized chart was used to retrieve patient data. The data were collected by two independent individuals who were blinded to the blood bactDNA results. The following data were collected: age, gender, body mass index (BMI), duration of disease, previous Crohn’s disease-related surgery, C-reactive protein (CRP) level, white blood count, serum albumin, disease location, preoperative drug use (immunosuppressive drugs), Crohn’s Disease Activity Index (CDAI) before surgery,23 indication for surgery, status of surgery (elective or emergency), laparoscopic surgery, smoking status, number of anastomoses, American Society of Anesthesiologists (ASA) wound classification, operative time, postoperative glucose value, and information regarding postoperative outcome. The blood glucose measurements were performed 48 h postoperatively and hyperglycemia was defined as glucose values more than 125 mg/dL.24
2.3. Detection of bactDNA fragments in blood BactDNA in blood samples was identified by conducting a broadrange PCR and then sequencing a conserved region of the 16S ribosomal RNA (16SrRNA) gene, as previously described.18–20,22 Blood sample DNA was extracted using the QIAamp DNA Mini Kit (Qiagen, 51304) according to the manufacturer’s instructions. The yield and purity of DNA were measured by reading A260 and A280 values in a Shimadzu spectrophotometer. The isolated DNA was used as a template to amplify the hypervariable V3 regions of the 16SrRNA gene. The universal PCR primers were V3F (5′-CCAGACTCCTACGGGAGGCAG-3′) and V3R (5′-CGTATTACCGCGGCTGCTG-3′). Positive and negative controls were evaluated in duplicate in each assay to avoid false-positive results. Genomic DNA extracted from the microbiota in ileum contents was added as a positive control. The negative control consisted of nuclease-free water (ABI, AM9930) and PCR mixture (without template). An aliquot of DNA (100 ng) was added to each reaction mixture. The PCR reactions were performed with a thermocycling program consisting of 35 cycles in a GeneAmp 9700 (Applied Biosystems, Foster City, CA) using the following protocol: 95°C for 40 s, 55°C for 40 s, and 72°C for 60 s. The resulting PCR amplicons were checked for the correct size (203 bp) on 2% agarose gels that were stained with ethidium bromide and visualized under ultraviolet light.
2.4. Postoperative outcomes We were particularly interested in the following 30-day postoperative outcomes: postoperative complications, time to bowel movement, postoperative blood transfusion requirement, duration of postoperative hospitalization, postoperative stay in the intensive care unit, readmission, and reoperation. Special attention was also paid to the organ dysfunction and systemic inflammatory response syndrome (SIRS).25 All complications were defined as those occurring within 30 days from the date of surgery. Early postoperative complications were divided into infectious and noninfectious complications. The infectious complications included wound infection, intra-abdominal abscess, anastomotic leak, pneumonia, urinary tract infection, and bacteremia. The noninfectious complications included postoperative ileus, small bowel obstructions, dehydration, and others.
2.5. Statistical analysis Continuous variables are presented as the mean ± SD and were compared using Student’s t-test for normally distributed variables. Fisher’s exact test or the χ2 test was used to compare categorical variables. The relative risks and 95% confidence intervals (CIs)
BactDNA and postoperative outcome in CD were also calculated to identify potential effects. Bivariate regression analysis was performed to determine the association between the presence of bactDNA and postoperative complications and to identify potential confounding variables associated with bactDNA translocation and postoperative outcomes. All covariates associated with postoperative morbidity were selected for possible inclusion in the corresponding multivariable models if the p-value for the bivariate association was less than 0.05. Multivariable models were then manually constructed in a forward stepwise manner. A p-value of less than 0.05 was considered statistically significant.
261 in blood versus 12.8% (10/78) in patients without bactDNA translocation (p = 0.023). There were wound infections in 8 of 29 (27.6%) patients who had bactDNA translocation compared with 7 of 78 (9.0%) patients who did not have bactDNA translocation (p = 0.025). There were no differences in the rates of intraabdominal abscess (p = 0.296), anastomotic leak (p = 0.617), pneumonia (p = 0.178), urinary tract infection (p = 0.060), or bacteremia (p = 0.470) between patients with and without bactDNA translocation.
3.4. Postoperative noninfectious complications
3. Results 3.1. Demographic and clinical characteristics at baseline A total of 107 patients with Crohn’s disease met the inclusion criteria and were included in this study. Twenty-nine patients (27.1%) showed the presence of bactDNA in their blood samples. BactDNA was not present in the blood of the other 78 patients (72.9%). In this study, none of the included patients presented a positive blood microbiological culture at inclusion. The mean age at surgery was 33.1 ± 11.7 years and the mean Crohn’s disease duration was 7.9 ± 4.7 years. Seventeen patients (15.9%) required a diverting stoma and there were 81 (75.7%) patients with one anastomosis. There were nine (8.4%) patients with more than one anastomosis. There were 18 (16.8%) patients who received emergency surgical intervention. The most common indication for surgery was stricturing/obstruction (82.2%), followed by medically refractory disease (17.8%). Before surgery, steroids were given in 42 (39.3%) cases and immunomodulators were used in 52 (48.6%) patients. There were 15 (14%) patients treated with anti-TNF agents. There was a higher mean level of preoperative CRP in patients with bactDNA in their blood samples (p = 0.001). There were no significant differences between the groups for the following variables: age, gender, duration of disease, previous Crohn’s disease-related abdominal surgery, BMI, laparoscopic surgery, preoperative concomitant treatment, indication for surgery, smoking status, urgent surgery, number of anastomoses, white blood count, albumin level, ASA score, wound classification, operative time, postoperative hyperglycemia, active disease with a CDAI of 150 or more, or Montreal classification (Table 1).
3.2. Early postoperative outcomes There were no postoperative deaths. There were 55 complications in 28 (26.2%) patients. Table 2 indicates that 28 complications were noted in 15 patients with bactDNA in their blood compared with 27 complications in 13 patients without bactDNA translocation (p 1 White blood count, ×109/L (mean ± SD) C-reactive protein, mg/L (mean ± SD) Albumin, g/L (mean ± SD) ASA score, n (%) ≥3
