Abstract
Background/Aim: Organ/space surgical site infections (SSIs) are critical complications of pancreaticoduodenectomy. We investigated the impact of the time between division of the common hepatic duct and completion of biliary reconstruction [bile exposure (BE) time] on the occurrence of post-pancreaticoduodenectomy organ/space SSI. Patients and Methods: Sixty-one patients who underwent pancreaticoduodenectomy were retrospectively studied. The impact of perioperative variables and BE time on organ/space SSI occurrence was analyzed. Results: Organ/space SSIs occurred in 17 patients (28%). Patients were divided into two groups according to BE time. The incidence of organ/space SSIs was significantly higher in the long BE time group than in the short BE time group (42% versus 13%, p=0.0127). Multivariate analysis revealed that long BE times [odds ratio (OR)=4.8; p=0.0240] and soft pancreatic texture (OR=16.5; p=0.0106) were independent risk factors for organ/space SSIs. Conclusion: Long BE time is a risk factor for post-pancreaticoduodenectomy organ/space SSIs. Shortening BE time may reduce organ/space SSI occurrence.
Despite advances in surgical techniques, pancreaticoduodenectomy is a complex procedure that carries a high risk of postoperative infectious complications, including surgical site infections (SSIs). During the last decade, SSIs occur in up to 30% of patients (1-7). Organ/space SSIs contribute significantly to increased rates of sepsis, prolonged hospital stays, readmission, and increased costs for patient care (8, 9).
Recent reports suggest that long operation time is associated with intraoperative bacterial contamination, especially via biliary pathogens, which can lead to post-pancreatico-duodenectomy organ/space SSIs (1, 2, 10). Although bile is usually sterile, and even when the bile duct shows no signs of microbial colonization (11), preoperative endoscopic retrograde cholangiopancreatography and biliary drainage can result in bile colonization (bacterobilia) at the time of surgery (12, 13). Additionally, while Enterococcus species are commonly isolated in intraoperative bile cultures (9, 10, 14), the use of antibiotics with anti-enterococcal spectrum is desirable for prophylactic antibiotics before pancreaticoduodenectomy in case of preoperative biliary stenting (10, 14).
Moreover, although post-pancreaticoduodenectomy organ/space SSIs may, in part, result from intraoperative bacterobilia, its clinical relevance and direct association with postoperative complications remain controversial (1, 2, 15-18). This study aimed to clarify the impact of the time between the division of common hepatic duct to completion of biliary reconstruction (BR) [bile exposure (BE) time] on the occurrence of post-pancreaticoduodenectomy organ/space SSIs.
Patients and Methods
Study design and operation procedure. We retrospectively reviewed patients who underwent pancreaticoduodenectomy for the treatment of benign and malignant peri-ampullary lesions at the Jikei Kashiwa Hospital (Kashiwa City, Japan) between January 2015 and October 2017. Preoperative biliary drainage (PBD) was performed for patients with obstructive jaundice. The internal drainage techniques used for these patients were endoscopic retrograde biliary drainage and endoscopic nasogastric biliary drainage. During surgery, bile juice was sampled and cultured when the common hepatic duct was transected. The texture of the pancreas (soft or hard) was determined by surgeon according to the intraoperative pancreas palpation. All patients underwent modified Child's reconstruction with subtotal stomach-preserving pancreaticoduodenectomy. At the end of the operation, the peritoneal cavity was irrigated using 7,000 ml of normal saline (7 washings of 1,000 ml each) with a clean suction apparatus. Prior to the lavage, surgical globes were changed.
Baseline characteristics of patients.
At the end of procedure, closed drains were placed anterior to pancreatico-jejunal anastomoses and dorsal to choledocho-jejunal anastomoses in all patients. Prophylactic antibiotics were administered routinely within the perioperative period. One gram of Cefmetazole (second-generation cephalosporin) was administered intravenously within 1 h before surgery and at approximately 3-h intervals intraoperatively. Cefmetazole administration was continued 3 times per day, for 48 h postoperatively.
The presence of SSIs, including incisional and organ/space SSIs, was determined according to the Centers for Disease Control and Prevention's national nosocomial infections surveillance system (19). BE time was defined as the time between common hepatic duct division and BR completion, which was obtained from each patient's time records in the operating room. BE time and other continuous variables were divided into two (long and short) groups according to the cut-off value obtained by Receiver Operating Characteristic (ROC) curve analysis for univariate and multivariate analysis. Patients provided written informed consent prior to undergoing surgery. This study was conducted in accordance with the Declaration of Helsinki and with the approval of the Ethics Committee of the Jikei University School of Medicine [approval no. 30-150 (9171)].
Statistical analysis. To evaluate the relationship between each categorized factor in terms of BE time, a univariable analysis was conducted using the χ2 test or Fisher's exact test, as appropriate. A logistic regression model and ROC curve analysis were used to assess the predictive characteristics of BE time with potential confounder variables. A p-value of <0.02 in the univariate analysis or a p-value of <0.05 in the logistic regression analysis was judged statistically significant. All statistical analyses were performed using Stat View-J 5.0 (Avacus Concepts Inc., Berkeley, CA, USA) and MedCalc 8.1 software (MedCalc Software, Mariakerke, Belgium).
Receiver operating characteristic (ROC) curve analysis predicted the value of bile exposure (BE)-time for occurrence of organ/space surgical site infection (SSI).
Results
A total of 61 consecutive patients who underwent open pancreaticoduodenectomy were included in this study. Patient characteristics and perioperative variables are listed in Table I. The median BE time was 157 min (range=76-366; mean= 174±69) and patients were divided into two groups; the long (≥153 min; n=31) and short (<152 min; n=30) BE time groups. Of the 61 patients, 17 (27.9%) developed organ/space SSIs. ROC curve analysis indicated BE time achieved considerably high sensitivity (82%) and specificity (59%), for the prediction of organ/space SSI occurrence (Area under the curve: 0.66; 95% confidence interval=0.52-0.77; p=0.0299; Figure 1).
There was no significant relation between patient's characteristics and organ/space SSI (Table II). On univariate analysis for perioperative variables, soft pancreatic texture (p=0.0120) and long BE times (p=0.0127) were significantly associated with the occurrence of organ/space (Table III). By multivariate analysis using these two variables, soft pancreatic texture [p=0.0106; Odds ratio (OR)=16.5) and long BE times (p=0.0240; OR=4.8) were independently related to the occurrence of organ/space SSIs (Table IV).
Discussion
The development of post-pancreaticoduodenectomy SSIs is a common and serious complication. Studies have shown that prolonged operation times result in increased wound susceptibility to bacterial infection, thereby causing the development of organ/space SSIs (2, 20, 21). Studies have also reported that PBD increases the incidence rate of biliary infection and organ/space SSIs (15, 17, 22, 23). In fact, biliary pathogens match with those identified in the abdominal abscesses of 50-100% of patients who have undergone post-PBD pancreaticoduodenectomy (15, 24-26). While PBD is often associated with polymicrobial infections containing resistant microorganisms (27), prophylactic antibiotics cover only 30% of these biliary pathogens (28). Given these studies, we hypothesized that the increased chance of bile contamination in the abdominal cavity would also increase the incidence of SSIs, especially organ/space SSIs.
Comparison of the occurrence of organ/space surgical site infection (SSI) and patient characteristics.
Comparison of the occurrence of organ/space surgical site infection (SSI) and perioperative findings.
In the present study, BE time was newly defined as an index of bile exposure to the peritoneal cavity during pancreaticoduodenectomy. To our knowledge, this is the first study that focuses on the relationship between BE time during pancreaticoduodenectomy and postoperative occurrence of organ/space SSI. Regarding the texture of the pancreas, soft pancreas has been a known risk factor for organ/space SSIs (7, 17-19, 29-31). However, it was interesting that even after excluding the effects of soft pancreatic textures, BE time remained an independent factor. This indicates that shortening BE time is expected to decrease organ/space SSI occurrence, regardless of pancreatic texture. In the current study, cut-off value of BE time (153 min) was optimal and our approach for predicting organ/space SSI occurrence was very practical.
Independent factors for organ/space surgical site infection (SSI) occurrence as determined via multivariate analysis.
The bacterial contamination of the abdominal cavity at the end of surgery is related to the subsequent development of post-pancreaticoduodenectomy SSIs (1). Large-volume peritoneal lavage has been recommended to remove bacteria in order to prevent SSI (32). These reports suggest that making efforts in reducing bacterial contamination during surgery is effective in preventing organ/space SSI development. Based on the current study, reduction of BE time may reduce the chances of organ/space SSIs by reducing bile contamination. Performing biliary reconstruction promptly after bile duct division during pancreaticoduodenectomy is a reasonable and practical method for reducing the chance of biliary infection. Additionally, whether large-volume peritoneal lavage at the end of surgery can reduce the incidence of bacterial contamination is an issue that may warrant further investigation. The limitations of the current study, include a small sample size, retrospective study, and a single center study.
In conclusion, long BE time is an independent risk factor for post-pancreaticoduodenectomy organ/space SSIs, and shortening BE time may reduce organ/space SSI occurrence.
Acknowledgements
The Authors would like to thank Editage (www.editage.jp) for English language editing.
Footnotes
Author's Contributions
YK, SF, YI, KF and KY participated in literature research and drafting the manuscript. SF, HK, TM, YK, TH and TM participated in treating patients. TM and YK participated in analyzing the study data. All authors have read and approved of the final manuscript.
This article is freely accessible online.
Conflicts of Interest
The authors declare that there are no conflicts of interest regarding this study.
- Received June 5, 2019.
- Revision received June 28, 2019.
- Accepted July 4, 2019.
- Copyright© 2019, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.