Abstract
Our aim was to examine whether systemic inflammatory response (SIR) is associated with perioperative central venous catheter-related bloodstream infection (CVC-RBSI) in patients undergoing surgery for colorectal cancer (CRC) with administration of parenteral nutrition (PN). Between May 2002 and August 2009, 310 patients undergoing CRC surgery were enrolled. A CVC was inserted in all patients for administration of PN. Among these patients, 117 developed fever, and blood cultures for diagnosis of CVC-RBSI were obtained from them. A final total of 22 patients were diagnosed as having CVC-RBSI. Univariate analysis was performed to evaluate the risk factors for CVC-RBSI using the clinical characteristics of the patients. The results of univariate analyses revealed that only the Glasgow Prognostic Score was an independent risk factor for CVC-RBSI (odds ratio 3.733; 95% confidence interval 1.416-9.844; p=0.008). The Glasgow Prognostic Score is associated with CVC-RBSI in patients undergoing CRC surgery with administration of PN.
- Central venous catheter
- central venous catheter-related bloodstream infection
- Glasgow Prognostic Score
- parenteral nutrition
- systemic inflammatory response
Use of a central venous catheter (CVC) plays a crucial role in the peri-operative management of patients with colorectal disease. Because most patients with colorectal cancer (CRC) have insufficient oral food intake because of bowel obstruction, parenteral nutrition (PN) via a CVC is required to provide them not only with nutritional support but also with fluid supplementation. Despite the development of antibiotics (1) and infection-control devices (2), once CVC-related bloodstream infection (CVC-RBSI) occurs, it can be lifethreatening (3).
A number of studies have revealed that the presence of a systemic inflammatory response (SIR) is associated with poor outcome in patients undergoing surgery for advanced cancer (4, 5). Because SIR is thought to be due to infiltration of pro-inflammatory lymphocytes, cytokines and chemokines in the tumor microenvironment, which predisposes the tumor to further progression, invasion and metastasis, it reflects the tumor versus host interaction (6). In addition, there is increasing evidence that the Glasgow Prognostic Score (GPS) (7, 8), an inflammation-based prognostic score that includes only serum C-reactive protein (CRP) and albumin, is one of the most useful scoring systems for prognostication of patients with several kinds of cancers (9-11).
Although mostly previous studies have used the GPS to investigate the relationship between advanced cancer and patient outcome, a recent investigation has revealed a close relationship between the modified GPS (mGPS) and infectious complications in patients undergoing curative resection for CRC (12). Since there have been few studies of the GPS in relation to CVC-RBSI in patients with CRC receiving PN via a CVC, we hypothesized that patients with SIR might have a more pronounced tendency for infectious conditions than patients without SIR. Therefore, the aim of the present study was to evaluate SIR, including the GPS, for prediction of peri-operative CVC-RBSI in patients undergoing surgery for CRC with administration of PN.
Patients and Methods
We retrospectively collected data for 310 patients (male:female= 193:106), who underwent CVC insertion for CRC surgery at the Department of Gastroenterological Surgery, Dokkyo Medical University Hospital, between May 2002 and August 2009, under the care of the same trained surgical team and nursing staff. Because all patients undergoing CRC surgery received the same treatment, based on a defined clinical path, they also received the same PN regimen via the CVC. Basically, the clinical path stipulated CVC insertion on the day before surgery in patients who had sufficient oral intake. However, most patients for whom oral intake was not possible, or for whom restriction of oral intake was necessary because of bowel obstruction, underwent CVC insertion on the day of admission, and were administered PN in preparation for bowel cleaning and nutritional support.
Routine laboratory measurements including the serum levels of CRP, albumin and tumor markers, such as carcinoembryonic antigen (CEA) (upper physiological value 5 ng/ml) (13) and carbohydrate antigen 19-9 (CA19-9) (upper physiological value 37 U/ml) (14), were carried out on the day of admission in order to exclude any effects attributable to inflammation associated with sequential preoperative examinations. None of the patients had clinical evidence of infection or other inflammatory conditions, and none had undergone pre-operative chemotherapy or irradiation.
GPS was estimated as described previously. Briefly, patients with both an elevated CRP level (>1.0 mg/dl) and hypoalbuminemia (<3.5 g/dl) were allocated a score of 2. Patients in whom only one of these biochemical abnormalities was present were allocated a score of 1, and those in whom neither of these abnormalities was present were allocated a score of 0 (15).
Interventions. All CVCs were inserted by the single-puncture method in the patient room using the maximal barrier precaution technique (16). The insertion area was disinfected with 10% povidone-iodine or 0.05% chlorhexidine (17) and draped. Observation with ultrasonography was routinely performed before the procedure to clarify the relationship between the vein and artery or other organs. However, ultrasonography-guided CVC insertion was not performed routinely. Two types of CVC were used for CVC insertion. One was a Groshong® catheter (Groshong, 4.0-French single-lumen 60-cm catheter, Bard Access Systems, Inc., Salt Lake City, UT, USA) and the other was an Argyle® catheter (Argyle, 16-gauge single-lumen 30-cm catheter, Nippon Sherwood, Tokyo, Japan) (18, 19). CVCs were inserted via the external jugular vein, internal jugular vein, subclavian vein, peripheral veins of the upper extremities or the femoral vein. Topical anesthetic skin infiltration with 1% lidocaine was performed except for insertion of the CVC via the external jugular vein and veins of the upper extremities. Accurate catheter tip placement was confirmed by chest- or abdominal X-ray film. No procedures were performed under intravenous sedation. Malposition of the catheter tip, conversion of the insertion site, oozing or hematoma formation at the insertion point, arterial bleeding, and symptoms of nerve injury were clarified as complications of CVC insertion. Proximal CVCs were defined as CVCs that were inserted via the internal jugular vein, external jugular vein (20) or subclavian vein, distal CVCs were defined as CVCs that were inserted via the femoral vein and the peripheral veins of the upper extremities.
Maintenance. After CVC insertion, all the patients were followed-up with routine route exchange every three days until CVC removal. CVCs were removed whenever fever occurred or if symptoms of infection were present, such as skin redness and pus discharge at the insertion point, and routine blood culture was performed for diagnosis of CVC-RBSI whenever the patient had a fever spike (>38°C). Routine CVC tip culture or blood culture was not performed in other cases. The period from catheter insertion to removal was defined in terms of catheter-days (21).
Definition of CVC-RBSI. In accordance with the Centers for Disease Control and Prevention guidelines (CDC guidelines) (http://www.cdc.gov/mmwr/preview/mmwrhtml/rr5110a1.htm), CVC-RBSI was diagnosed according to at least one of the following criteria. Criterion 1: The patient had a recognized pathogen cultured from one or more blood cultures, and the cultured pathogen was not related to any infection at another site. Criterion 2: The patient had at least one of the following signs or symptoms: fever (>38°C), chills, or hypotension. On the basis of the CDC guidelines, we considered that CVC-RBSI had developed if blood cultures from patients with fever were positive (21).
Definition of operative curability. On the basis of the Japanese Classification of Colorectal Carcinoma (Japanese Society for Cancer of the Colon and Rectum, Second English Edition) (22), residual tumor is diagnosed as: R0, no residual tumor; R1, no residual tumor, but tumor at the resection margin suspected; R2, macroscopically evident residual tumor. On the basis of this definition, operative curability is defined as: curability A (Cur A), R0 in TNM stage I, II or III; curability B (Cur B), R0 in TNM stage IV or R1 in any TNM stage; curability C (Cur C), R2 in any TNM stage.
Statistical analysis. Data are presented as the mean±standard deviation (SD) and 95% confidence interval (CI). Differences in values among the three groups were analyzed using the Kruskal-Wallis test. Odds ratios with 95% CIs were calculated using univariate analysis, which was performed using logistic regression analysis. Statistical analyses were performed using the SPSS statistical software package, version 16.0 (SPSS Inc., Chicago, IL, USA), at a significance level of p<0.05.
Results
Three hundred and ten patients had a total of 5,702 catheter-days. The mean age of the patients was 67.2 years, the mean CVC insertion length was 17.8 cm, and the mean period of CVC insertion was 19.1 days. The overall CVC-RBSI rate was 3.85 per 1,000 catheter-days.
Table I shows the relationships between the background characteristics and GPS in patients undergoing CRC surgery with administration of PN. There were no significant differences in the background characteristics, except for TNM staging (p<0.001); operative curability (p<0.001); fever (presence/absence) (p=0.001) and blood culture positivity (presence/absence) (p=0.009) among the three groups classified by GPS. Fever (>38°C) occurred in 117 patients (37.7%). Among the patients with fever, 22 (18.8%) were diagnosed as having CVC-RBSI on the basis of blood culture positivity. There was no significant difference in the period of CVC insertion between the two groups with (n=22, 20±11 days) and without (n=288, 19±19 days) CVC-RBSI. Among the 22 patients with CVC-RBSI, five were diagnosed as having had CVC-RBSI before surgery. Moreover, there was no significant difference in the period of CVC insertion between the two groups in which CVC-RBSI was diagnosed before (n=5, 22±10 days) and after (n=17, 19±11 days) surgery.
The results of blood culture revealed positivity for Staphylococcus in 14 patients (14/22, 63.6%) (GPS 0: 7; GPS 1: 1; GPS 2: 6), Bacteroides in 3 patients (GPS0: 3), and Bacillus (GPS 1: 1), Haemophilus (GPS 0: 1), Candida (GPS 0: 1), Pseudomonas (GPS 2: 1) and Escherichia coli (GPS 1: 1) in one patient each, respectively. Most pathogens were detected in patients with GPS 0 or 1, except for Staphylococcus (GPS 2: 6) and Pseudomonas (GPS 2: 1).
Relationships between background characteristics and Glasgow Prognostic Score (GPS) in patients undergoing colorectal cancer surgery with administration of parenteral nutrition.
Similarly, there were no significant differences in the period of CVC insertion, length of the inserted part of the CVC or body mass index (BMI) among the three groups. There were significant differences in age (p=0.004), white blood cell (WBC) count (p<0.001), neutrophil ratio (p<0.001), and the serum levels of CRP (p<0.001), albumin (p<0.001), CEA (p<0.001) and CA19-9 (p=0.019) among the three groups (Table II).
Relationships between clinical characteristics and Glasgow Prognostic Score in patients undergoing colorectal cancer surgery with administration of parenteral nutrition. Data are the mean±SD.
Univariate analysis was performed to evaluate the relationship between the presence of perioperative CVC-RBSI and clinical characteristics including sex; age; period of CVC insertion; length of the inserted part of the CVC; incidence of insertion; site of insertion; type of CVC catheter; type of disinfectant; WBC count; neutrophil ratio; serum levels of CRP; albumin; CEA and CA19-9; BMI, tumor site, operative curability, TNM staging and GPS. This demonstrated that only GPS was associated with CVC-RBSI (odds ratio=3.733; 95% CI=1.416-9.844; p=0.008) (Table III).
Discussion
Because there were no significant intergroup differences in background characteristics, except for TNM staging, operative curability, fever and blood culture positivity, among the three groups classified by GPS, there was rigid uniformity among the patients, thus providing strong validity for this retrospective study. With regard to clinical characteristics that were excluded, because GPS is thought to reflect hypercytokinemia due to tumor versus host interaction, it is acceptable that GPS would show close relationships to TNM staging and operative curability C, as well as high levels of serum CEA and CA19-9 due to tumor growth and progression.
Similarly, because the immune system may be suppressed by such hypercytokinemia due to SIR, it is well accepted that there are also close relationships between GPS and factors indicative of infection, such as fever and blood culture positivity, as well as an increased WBC count and neutrophil ratio.
On the other hand, although we cannot fully explain why older patients have a higher GPS in this retrospective study, other CVC-related characteristics, such as the type of catheter, type of disinfectant, site of insertion, incidence of insertion, period of insertion and catheter insertion length, were not associated with GPS.
Univariate analyses of clinical characteristics in relation to central venous catheter-related bloodstream infection in patients undergoing colorectal cancer surgery with administration of parenteral nutrition.
Therefore, the results of univariate analyses using clinical characteristics clearly demonstrated that pre-operative SIR, as evidenced by GPS, was an independent risk factor for CVC-RBSI among the clinical characteristics of patients undergoing surgery for CRC.
It is well-known that the main causes of CVC-RBSI are related to the external environment of blood vessels, such as contamination via connectors along the insertion route, dressing change at the insertion point, administration of PN or surgery, because the results of blood cultures revealed that Staphylococcus, which is commonly distributed on the body surface as part of the normal flora, was a major pathogen responsible for CVC-RBSI, whereas only a few patients had pathogens associated with colorectal surgery, such as Bacteroides (3/22; 13.6%) and Escherichia coli (1/22; 4.5%).
Moreover, although most pathogens were detected in patients with GPS 0 or 1, Staphylococcus (GPS 2: 6/14, 43%) and Pseudomonas (GPS 2: 1/1, 100%) had a higher incidence in patients with GPS 2 than did the other pathogens. Similarly, there was a significant difference among the three groups in the ratio of blood culture positivity (p=0.009). The GPS 2 group had a higher incidence of blood culture positivity (7/39, 18%) than the GPS 0 (12/172, 7.0%) and GPS 1 (3/99, 3.0%) groups. Therefore, patients with GPS 2 have a higher proportion of infectious conditions in comparison with those with GPS 0 or 1.
Recently, Moyes et al. (12) demonstrated that the presence of SIR, as evidenced by the modified Glasgow prognostic score (mGPS) (8), predicts an increased incidence of post-operative infectious complications in patients undergoing potentially curative resection for CRC. Although the relationship between an elevated mGPS and post-operative infections is not clear, they suggested the following reasons for this phenomenon. Firstly, an elevated mGPS may reflect compromised cell-mediated immunity, as the level of CRP is associated with lymphopenia (23, 24) and an impaired T-lymphocytic response (6) in patients with CRC. Secondly, CRP is also associated with components of the innate immune system, including complement, and this response may also be compromised (25, 26). Thirdly, a mGPS of 2 reflects loss of lean tissue and protein (27, 28), which is likely to further compromise immune function. On the basis of the above reasons, they concluded that compromised immune function may be present before surgery, and may influence post-operative infectious conditions (12).
Similarly, several studies have provided concrete evidence that not only is a high level of serum CRP (29, 30) associated with infectious conditions, but so is a low level of serum albumin (31, 32). Because interleukin-6 (IL-6) not only down-regulates albumin production (33) but also up-regulates CRP production in hepatocytes (34, 35), these adverse effects on infectious conditions are based on the same phenomenon as hypercytokinemia associated with a high serum level of IL-6 (36). In fact, a high serum IL-6 levels have a much closer association with infectious conditions than do a high serum levels of CRP (37). Therefore, it is well acceptable that a GPS of 2, which reflects a high level of CRP and a low level of albumin, is a strong predictor of both peri-operative CVC-RBSI and post-operative infection.
Recent advances in molecular biological technology for production of antibody agents have led to the development of an innovative therapeutic drug, tocilizumab, a recombinant humanized antiantibody against human IL-6 receptor, for various chronic autoimmune inflammatory diseases, such as rheumatoid arthritis. In practice, blocking of the IL-6 signaling pathway with tocilizumab has dramatically improved all the signs and symptoms of such diseases (38). Although there is no indication that such antibody agents would be able to prevent peri-operative complications against a background of hypercytokinemia, an optional treatment for future consideration would be to reduce CVC-RBSI in patients with infectious conditions due to elevation of the serum IL-6 level, especially in patients with CRC and GPS of 2.
Accordingly, estimation of the GPS would not only be clinically useful for identifying patients at high risk of developing CVC-RBSI, but also simple to perform before surgery in any setting.
Footnotes
-
Conflicts of Interest
We have no conflicts of interest to declare. We received no funding/grant support for this study.
- Received June 15, 2012.
- Revision received August 6, 2012.
- Accepted August 8, 2012.
- Copyright© 2012 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.