JAC Advance Access originally published online on February 28, 2007
Journal of Antimicrobial Chemotherapy 2007 59(5):964-970; doi:10.1093/jac/dkm028
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Duration of antimicrobial prophylaxis in patients undergoing hepatectomy: a prospective randomized controlled trial using flomoxef
Department of Gastroenterological Surgery, Yokohama City University Graduate School of Medicine, Yokohama, Japan
* Corresponding author. Tel: +81-45-787-2650; Fax: +81-45-782-9161; E-mail: togo{at}med.yokohama-cu.ac.jp
Received 14 September 2006; returned 28 October 2006; revised 18 January 2007; accepted 19 January 2007
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and methodsResultsDiscussionTransparency declarationsReferences |
|---|
Objectives: Although the usefulness of antimicrobial prophylaxis for clean-contaminated surgery has been recognized, only a few randomized controlled studies on the duration of administration after hepatectomy have been performed. We investigated the duration of antimicrobial prophylaxis after hepatectomy.
Methods: The subjects were 180 patients who underwent hepatectomy without reconstruction of the biliary or intestinal tract between April 2003 and March 2006 at our department. The patients were randomly allocated to groups to be treated with flomoxef sodium as antimicrobial prophylaxis for 2 days (89 patients) or 5 days (91 patients), including the operation day. The presence or absence of systemic inflammatory response syndrome (SIRS) and infections was investigated.
Results: No significant differences were noted in patient background between the two groups. Infections occurred in seven and six patients in the 2 day and 5 day treatment groups (7.9% and 6.6%), respectively, showing no significant difference between the two groups. No significant difference was noted when the cases were divided into surgical site infections and remote infections. The positive rate of SIRS was significantly higher in the 2 day treatment group than in the 5 day treatment group on days 2 and 3 after surgery. The risk factors in patients who developed infections were blood loss, operation time and the complication of biliary fistula.
Conclusions: Two day administration of flomoxef sodium may be sufficient for antimicrobial prophylaxis after hepatectomy. However, when SIRS is positive on post-operative day 2, and induction of liver failure is of concern, it may be safer to continue antimicrobial drug administration until SIRS is eliminated.
Keywords: liver resection , post-operative infection , systemic inflammatory response syndrome , SIRS
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionTransparency declarationsReferences |
|---|
Hepatectomy is classified into clean-contaminated surgery because the bile duct is dissected. The usefulness of antimicrobial prophylaxis for general clean-contaminated surgery has been confirmed by randomized controlled studies.1,2 For hepatectomy, antimicrobial prophylaxis may be essential because impairment of glucose tolerance due to reduced liver function, reduced immunity, relatively long operation time and relatively large blood loss are present. Thus, patients undergoing hepatectomy are a group at high risk of surgical site infection (SSI),3 in addition to the fact that this operation is a clean-contaminated one. However, only two randomized controlled trials (RCTs) on the usefulness of post-operative antimicrobial prophylaxis after hepatectomy have been performed,4,5 and the results were contradictory. Under such circumstances, the duration of post-operative administration of antibiotic drugs is decided on the basis of experience at each facility.
The American Thoracic Society and the Society of Critical Care Medicine proposed a new concept, systemic inflammatory response syndrome (SIRS), in 1991.6 SIRS is a condition in which a systemic inflammatory reaction is induced by stress such as surgery, and its main feature is considered to be hypercytokinaemia. This high-cytokine condition is involved in homeostasis maintenance in the body, but it may impair organs when it is protracted or excessive.7 In the field of gastrointestinal surgery, SIRS has been attracting attention as an index of surgical stress early after surgery and a warning sign of complications.8
We performed a prospective RCT. The patients were divided into groups treated with flomoxef sodium (product name: Flumarin), which appeared to have been most useful in previous studies,9,10 as post-operative antimicrobial prophylaxis for 2 (operation day and the following day) or 5 days (until post-operative day 4). The incidence of post-operative infections [SSI and remote infections (RI)] and the positive rate of SIRS were investigated.
|
Patients and methods |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionTransparency declarationsReferences |
|---|
Patients
In our university hospital, patients who underwent elective liver resection for hepatic lesions were enrolled in this study. As selection criteria, patients who were concomitantly treated with the resection of other organs—resection and anastomosis of the bile duct; drainage and resection and anastomosis of the digestive tract—and patients in whom apparent infection was noted at the time of surgery were excluded. Patients with a past medical history of allergic reaction to ß-lactams, a pre-operative serum creatinine level of 
1.5 mg/dL, diabetes mellitus under insulin control, treatment with antibiotics within 1 week before surgery, and being judged ineligible by the attending physician were also excluded. Patients who did not consent to this study were excluded. Informed consent was obtained from each patient before surgery.
Stratification, randomization and blinding
Before surgery, patients who met the inclusion criteria were randomly allocated to the 2 day or 5 day treatment groups by the minimization method. The stratification factors were age (older than 60 years of age and younger than 60), diseases (primary liver cancer, metastatic liver cancer and others) and the planned resection rate (30% and <30%). Allocation was performed in a blind manner using a personal computer. This study was approved by the Ethics Committee of our university.
Flomoxef sodium was used as an antimicrobial prophylaxis to prevent post-operative infections. On the operation day, 1 g was administered 30 min before surgery, 1 g every 3 h during surgery, 1 g 2 h after the completion of surgery and then 2 g a day after the operation day (1 g every 12 h). The patients were randomly divided into the 2 day and 5 day treatment groups and were then treated with this regimen and compared. No patient received pre-operative administration of oral antimicrobial agents.
The primary endpoint of this study was the incidence of SSI. Secondary endpoints were the positive rate of SIRS and other complications such as RI, bile leakage and diarrhoea.
Haematological and biochemical tests were performed before surgery, immediately after surgery and 1, 2, 3, 5, 7, 10 and 14 days after surgery, and additional tests were performed as needed. Routine bacteriological tests were performed after drain removal, after catheter removal from the central vein, or of sputum after removal of an intratracheal tube when the tubes were not removed immediately after surgery, and additional tests were performed as needed. Imaging examinations, such as plain chest radiography, abdominal CT scanning and abdominal ultrasonography, were performed as needed when post-operative infection was suspected.
Judgement criteria of infections and SIRS
The incidence of infections (infections of the surgical site and distant organs) 30 days after surgery and the positive rate of SIRS were compared between the two groups. As a rule, infections were diagnosed when the culture test of the site suspected as being infected was positive between immediately after surgery and 30 days following surgery. For SSI, the Guidelines for Prevention of Surgical Site Infection established by the CDC were used.11 SSI includes superficial, deep incisional and organ/space SSI and was diagnosed on the basis of the presence of the following findings within 30 days after surgery: inflammatory findings, such as fever and flare, drainage of pus from the incision or drain, detection of a pathogen by culture of fluid or tissue sample and fluid retention on imaging indicating the presence of pus in a deep region. Cases judged as SSI by physicians were also included even though pus drainage and pathogens on culture were absent. For example, when fever and right hypochondrial pain developed after drainage removal following surgery and exudate retention was noted on the cut surface of the liver on CT or echo, the case was diagnosed as infection of the cut liver surface, although nothing abnormal was found in the culture.
For SIRS, the consensus criteria established by the American Thoracic Society and the Society of Critical Care Medicine were used.6 When two of the following four conditions were present, the case was diagnosed as SIRS: (i) body temperature of 
36°C and 38°C; (ii) a pulse rate of 90/min; (iii) a respiratory rate of 20/min or PaCO2 > 32 torr; and (iv) white blood cell counts of 12 000/mm3 and 4000/mm3 or 10% immature cells.
Diagnosis of SSI and RI was made by the attending physician and a physician in the infectious control team of our university.
When post-operative infection was diagnosed during the study, the treatment was changed to therapeutic antimicrobial agents.
Diagnostic criteria of bile leakage
Diagnosis of biliary complications was based on one or more of the following post-operative findings: (i) drainage of bile from the abdominal wound and drain, showing a total bilirubin level of >5 mg/mL or three times the serum level in the discharge fluid; (ii) an intra-abdominal accumulation of bile confirmed by percutaneous drainage; and (iii) cholangiographic evidence of bile leakage.12
Our department had previously administered antimicrobial prophylaxis for 5 days after hepatectomy, and the incidence of post-operative SSI was 9.4% in the last 3 years. This study was a non-inferiority study to demonstrate the equivalence of the 2 day and 5 day treatment groups.13
When the incidence was predicted to be 6% in consideration of a decrease in the SSI incidence due to various countermeasures against it during the study, the number of cases required to analyse non-inferiority was 89 per group, when the significance level, power and tolerance difference were set to 0.025%, 0.8% and 10%, respectively. Thus, 89 cases per group were targeted.
All analyses were performed as intention-to-treat. For inter-group comparison of the incidence of post-operative infections, the non-inferiority test with a non-inferiority margin of 7% was performed, and P < 0.05 was regarded as significant.13 For inter-group comparison of items other than the incidence of post-operative infections, the Mann–Whitney U test and 
2 test were used. P < 0.05 was regarded as significant.
The significance of differences was analysed every 6 months to decide upon discontinuation of the study from an ethical viewpoint when the incidence of infections was significantly higher in one of the two groups.
|
Results |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionTransparency declarationsReferences |
|---|
Number of entries
Of 196 patients who underwent hepatectomy at our department between April 2003 and March 2006, 180 patients who met the inclusion criteria were enrolled in the study. Eighty-nine and 91 patients were allocated to the 2 day and 5 day treatment groups, respectively (Figure 1). The reasons for exclusion of 16 patients were as follows: treatment with antimicrobial agents before surgery in 2; reconstruction of the intestine or biliary tract necessary during surgery in 2; diabetic complication treated with insulin in 3; renal dysfunction in 2 and withholding of consent in 7. After allocation, no patients were excluded from the study.
|
Background characteristics of the patients
The indications for liver resection and background patient characteristics are shown in Table 1. There were no significant differences in tumour pathology or in any category between the two groups.
|
Surgical procedures and intraoperative results
There were no significant differences in any category between the two groups (Table 2).
|
Post-operative course
Infectious complication rate. The incidence of post-operative infections was seven (7.9%) and six (6.6%) in the 2 day and 5 day treatment groups, respectively, revealing non-inferiority. When post-operative infections were divided into SSI and RI, the SSI incidence was four (4.5%) and four (4.4%) in the 2 day and 5 day treatment groups, respectively, showing non-inferiority. In contrast, the RI incidence was three (3.4%) and two (2.2%), respectively, so that there was a failure to demonstrate non-inferiority (Table 3).
|
Infected site. Infections occurred in 13 patients, accounting for 7.2% of all patients. SSI occurred in eight: intra-abdominal infection in five and wound infection in three. RI occurred in five: catheter infection in four and pneumonia in one. The causative bacteria were Gram-positive cocci in three and Gram-negative bacilli in two in the 2 day treatment group and Gram-positive cocci in all cases in the 5 day treatment group (Table 3).
Positive rate of SIRS. The positive rates of SIRS on days 2 and 3 after surgery were 31.7% (26 cases) and 14.6%, respectively, in the 2 day treatment group and 14.1% and 4.7%, respectively, in the 5 day treatment group, showing that the positive rate was significantly higher in the 2 day treatment group on days 2 and 3 after surgery (Table 4). However, no significant difference was noted between the two groups after post-operative day 4 (Figure 2). The cumulative numbers of days with SIRS after surgery were 1.34 and 0.95 in the 2 day and 5 day treatment groups, respectively, showing a slightly longer duration in the 2 day treatment group, although the difference was not significant (Table 4).
|
|
Serious complication, diarrhoea and hospital stay after operation. There was no post-operative death during the admission period in either group. No severe complications were observed. If complications were defined as conditions in which the admission period after surgery exceeded 22 days (mean + SD), the number of patients with complications was 18 in the 2 day group (20.2%) and 14 in the 5 day group (15.4%), the difference being not significant. Table 5 shows the complications. All patients recovered and were discharged from the hospital. Diarrhoea was noted in two patients in the 2 day group alone, but Clostridium difficile was not detected in these patients. The number of patients in whom SIRS was positive for 4 days or longer and imipenem/cilastatin was administered was 12 in the 2 day group and 9 in the 5 day group, the difference being not significant.
|
Risk factor of post-operative infections. On comparison of patients who developed infections with those who did not, the operation time was significantly longer in the former group, and the blood loss and bile fistula complication rate were also higher (Table 6).
|
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionTransparency declarationsReferences |
|---|
Since hepatectomy is invasive to the liver, which is important for defence against infection, and is performed in patients with reduced liver function, such as those with hepatic cirrhosis, patients may readily progress to liver failure upon intra-abdominal infection.3 Thus, prevention and treatment of post-operative infection are important to improve the outcomes of hepatectomy.
Only a few randomized controlled studies on the duration of antimicrobial prophylaxis after hepatectomy have been performed. Wu et al.4 performed a comparative study of groups untreated and treated for 7 days with cefazolin and gentamicin and found that antimicrobial prophylaxis did not prevent post-operative infection. In contrast, Sano et al. compared groups treated with cefazolin only during surgery and treated until post-operative day 5 and found that antimicrobial prophylaxis was useful for the prevention of post-operative infection and signs of infection, insisting on the effectiveness of post-operative administration. Thus, the reports were contradictory. In this study, we performed an RCT of the duration of administration using flomoxef sodium in patients who underwent hepatectomy without reconstruction of the intestine or the biliary tract.
The target-contaminating bacteria in the field of gastrointestinal surgery are the air-borne bacteria, Staphylococcus aureus, enteric bacteria, Escherichia coli and Klebsiella pneumoniae and an anaerobe, Bacteroides fragilis. The antibacterial activity of flomoxef sodium against these bacteria is superior to those of cefazolin and cefmetazole,14 and the drug causes fewer adverse events, similar to other cephem antibiotics, and has less effect on intestinal flora.15 In our experience in the last 10 years, the lowest incidence of post-operative infection was noted when flomoxef sodium was administered, and the rate of induction of multidrug-resistant bacteria was also low.10 Therefore, flomoxef sodium was considered the most adequate antimicrobial agent and was thus used for antimicrobial prophylaxis in this study.
Regarding how to administer antimicrobial prophylaxis, the CDC guidelines specify the maintenance of an effective blood level from the initiation of surgery until 2 h after closure of the abdomen.11 Accordingly, 1 g of flomoxef sodium was administered 30 min before surgery, followed by 1 g every 3 h. Pre-operative treatment with oral non-absorbable antimicrobial agents was useful in some reports. Condon and Wittmann16 recommended that a pre-operative protocol of bowel preparation and oral antibiotics should be administered, as a substantial serum level of erythromycin during operation may play some role in reducing the perioperative infection rate.16 However, in this study, we could not choose such a protocol.
Regarding the establishment of the duration of antimicrobial prophylaxis, the mean number of days with post-operative SIRS, including the operation day, was 1.62 ± 1.59 days in the last 3 years before this study. On the basis of this, the 2 day treatment group covering the SIRS-positive period to some extent and the 5 day treatment group, in which the duration was mean + 2 SD, mostly covering the entire SIRS-positive period were established. In the CDC guidelines,11 maintenance of the effective blood level from anaesthesia induction until the completion of surgery is emphasized, but no need for post-operative administration is described. Bratzler et al.17 reported that 60% of cases were treated for >24 h (e.g. drugs were administered until the day following surgery) for cardiovascular, orthopaedic, large intestinal and gynaecological surgeries in the USA, indicating that post-operative administration was considered useful in many studies. Therefore, we considered that administration on the operation day alone is not practical for clinical cases.
This randomized controlled study was a non-inferiority study to investigate equivalence. The sample size established first was 89 per group. SIRS-positive cases were to be carefully observed and immediately treated with therapeutic antimicrobial drugs in the case of infections, such as sepsis, and conditions that may deteriorate into liver failure, a serious complication.
No significant differences were noted in the incidence of RI or SSI between the 2 day and 5 day treatment groups, but SIRS was significantly inhibited on post-operative days 2 and 3 in the 5 day treatment group. These findings are similar to those reported by Sano et al.,5 who considered SIRS as a sign of suspected infection.
SIRS is determined by fever, pulse and respiratory rates, and by the white blood cell count, and is induced by overproduced humoral mediators including cytokines.9 Although it does not advance to cellular injury, SIRS is a dangerous condition that may lead to organ failure. Particular attention should be paid to septic SIRS, which occurs upon infection. According to Kitamura and Hirasawa,8 SIRS was noted in 65.6% of patients who entered an intensive care unit after gastrointestinal surgery, organ failure occurred in 4.9% of SIRS-positive cases, 6.3% died and the mortality rate increased as the SIRS-positive period was prolonged, showing that SIRS is an early warning sign of organ failure, and early countermeasures against SIRS should be taken. Therefore, SIRS-positive cases should be carefully observed, and when advancement to infections, such as sepsis, or liver failure is suspected, therapeutic antimicrobial drug administration should be immediately initiated.
Regarding the mechanism of SIRS inhibition by flomoxef sodium in this study, the inhibition of pathogen proliferation by flomoxef sodium may have prevented the enhancement of cytokine production by neutrophils.
When infection, particularly sepsis, develops after hepatectomy, fatal liver failure may occur.18–20 Thus, signs of infection should be carefully followed, and immediate treatment may be necessary. According to Sano et al., when a sign of infection is noted, antimicrobial drugs should be administered or changed, and not only blood sampling and various culture tests but also ultrasonography or CT may be necessary to evaluate possible intra-abdominal infection.5 Although the signs of infection described by Sano et al. are slightly different from those of SIRS,5 we also carefully observed SIRS-positive patients and administered imipenem/cilastatin as therapeutic antimicrobial drugs when the SIRS persisted for 4 days or longer, and no liver failure occurred in any patient.
Bratzler et al.17 reported that antimicrobial prophylaxis is administered for 24 h or longer in >60% of cases in the USA. Evidence for the optimal duration of administration may be needed in the future. Since no difference was noted in the incidence of post-operative infection between the 2 day and 5 day treatment groups, administration until post-operative day 1 may be sufficient for routine treatment, but additional administration after post-operative day 2 may be necessary for SIRS-positive cases, and SIRS may serve as an index for the duration of post-operative administration.
Two days may be a sufficient duration of flomoxef sodium administration because no significant differences were noted in the incidence of post-operative infection or SSI between the 2 day and 5 day treatment groups. However, when SIRS is positive on post-operative day 2, and induction of liver failure is of concern, it may be safer to continue antimicrobial drug administration until SIRS is eliminated.
|
Transparency declarations |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionTransparency declarationsReferences |
|---|
None to declare.
|
References |
|---|
|
TopAbstractIntroductionPatients and methodsResultsDiscussionTransparency declarationsReferences |
|---|
1 Baum ML, Anish DS, Chalmers TC, et al. (1981) A survey of clinical trials of antibiotic prophylaxis in colon surgery: evidence against further use of no-treatment controls. N Engl J Med 305:795–9.[Abstract]
2 Song F and Glenny AM. (1998) Antimicrobial prophylaxis in colorectal surgery: a systematic review of randomized controlled trials. Br J Surg 85:1232–41.[CrossRef][ISI][Medline]
3 Makuuchi M and Sano K. (2004) The surgical approach to HCC: our progress and results in Japan. Liver Transpl 10:S46–52.[CrossRef][Medline]
4 Wu CC, Yeh DC, Lin MC, et al. (1998) Prospective randomized trial of systemic antibiotics in patients undergoing liver resection. Br J Surg 85:489–93.[CrossRef][ISI][Medline]
5 Sano K, Takayama T, Makuuchi M. (2002) Prophylactic antibiotics in hepatectomy (in Japanese). Geka (Surgery) 64:1635–9.
6 Members of the American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference Committee. (1992) Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Crit Care Med 20:864–74.[ISI][Medline]
7 Ogawa M. (1998) Systemic inflammatory response syndrome — a concept for avoiding organ dysfunction induced by a "second attack". Surg Today 28:679–81.[ISI][Medline]
8 Kitamura N and Hirasawa H. (1995) A study on the outcome and pathophysiology of systemic inflammatory response syndrome (SIRS) in critically ill patients (in Japanese). JJAAM 6:662–72.
9 Tsuji T, Satoh H, Narisada M, et al. (1985) Synthesis and antibacterial activity of 6315-S, a new member of the oxacephem antibiotic. J Antibiot (Tokyo) 38:466–76.[Medline]
10 Togo S, Tanaka K, Sugita M, et al. (2005) Perioperative infection-control management for patients who underwent hepatectomy (in Japanese). Geka (Surgery) 67:178–84.
11 Mangram AJ, Horan TC, Pearson ML, et al. (1999) Guideline for prevention of surgical site infection, 1999. Infect Control Hosp Epidermiol 20:247–78.
12 Nagano Y, Togo S, Tanaka K, et al. (2003) Risk factors and management of bile leakage after hepatic resection. World J Surg 27:695–8.[CrossRef][ISI][Medline]
13 Hirotsu C. (1986) Some statistical problems in clinical trials. (1) Test for the equivalence of two drugs (in Japanese). Clin Eval 14:467–75.
14 Furukawa K, Onda M, Suzuki H, et al. (1999) The usefulness of conducting investigations on intra-abdominal bacterial contamination in digestive tract operations. Surg Today 29:701–6.[CrossRef][ISI][Medline]
15 Takesue Y, Yokoyama T, Akagi S, et al. (1999) Impact on intestinal flora after administration of flomoxef in gastroenterological surgery (in Japanese). Jpn J Chemother 47:74–9.
16 Condon RE and Wittmann DH. (1991) The use of antibiotics in general surgery. Curr Probl Surg 28:801–949.[ISI][Medline]
17
Bratzler DW, Houck PM, Richards C, et al. (2005) Use of antimicrobial prophylaxis for major surgery: baseline results from the National Surgical Infection Prevention Project. Arch Surg 140:174–82.
18 Yanaga K, Kanematsu T, Takenaka K, et al. (1986) Intraperitoneal septic complication after hepatectomy. Ann Surg 203:148–52.[ISI][Medline]
19 Anderson R, Saarela A, Tranberg KG, et al. (1990) Intraabdominal abscess formation after major liver resection. Acta Chir Scand 156:707–10.[ISI][Medline]
20 Takeda K, Togo S, Kunihiro O, et al. (2002) Clinicohistological features of liver failure after excessive hepatectomy. Hepatogastroenterology 49:354–8.[Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||