Influence of a Shorter Duration of Post-Operative Antibiotic Prophylaxis on Infectious Complications in Patients Undergoing Elective Liver Resection

Abstract

Background:

Antibiotic prophylaxis has been recommended to reduce post-operative infectious complications. Discontinuation of post-operative antibiotic administration within 24 hours of operation is currently recommended. Many surgeons, however, conventionally tend to extend the duration of prophylactic antibiotic use. In this study, we performed a retrospective analysis to assess the efficacy of extended post-operative antibiotic use in patients who underwent elective liver resection.

Patients and Methods:

A total of 208 consecutive patients who underwent liver resection without biliary reconstruction were investigated. Patients were divided into two groups according to the duration of post-operative antibiotic use: Only once after the operation (the post-operative day [POD] 0 group) and until three days after the operation (the POD 3 group). Post-operative complications in the two groups were analyzed and compared.

Results:

Incisional surgical site infections (SSIs) were observed in 5% of the POD 0 group and 3% of the POD 3 group (p = 0.517). Organ/space SSIs were observed in 2% of the POD 0 group and 3% of the POD 3 group (p = 0.694). Overall infectious complications including SSIs and remote site infections were observed in 12% of the POD 0 group and 11% of the POD 3 group. Multi-variable analyses revealed that the short-term post-operative antibiotic regimen did not confer additional risk for infectious complications.

Conclusions:

In elective liver resection, the administration of prophylactic antibiotics on the operative day alone appears to be sufficient, because no additional benefit in the incidence of post-operative infectious complications was conferred on patients given antibiotic agents for three days.

As a result of recent advances in surgical techniques and peri-operative management, liver resection has become a safer operation [1 –3]. Critical complications still develop in a few patients who undergo liver resection, however. Bile leakage and surgical site infections (SSIs), including both incisional and organ/space SSIs, are common causes of major morbidity after liver resection and lead to increased medical costs, longer hospitalization, and a reduced quality of life in the post-operative course [4 –7].

In its 1999 guideline, antibiotic prophylaxis was recommended by the Centers for Disease Control and Prevention (CDC) to reduce post-operative infectious complications [8]. In subsequent studies, the timing of prophylactic antibiotic administration was discussed, and the pre-operative prophylactic antibiotic is generally recommended to be administered within 60 minutes before skin incision and to be re-dosed during prolonged operations based on the antimicrobial half-life of the antibiotic or the amount of blood loss and fluids administered [9,10], and then stopped within 24 hours after the surgical procedure [8,11,12]. Post-operative antibiotic prophylaxis has been reported as effective after a digestive operation that is classified as a clean-contaminated surgical procedure [13,14].

Four randomized controlled trials (RCTs) on the efficacy of post-operative antibiotic prophylaxis after liver resection have been reported [15 –18]. Sano et al. [15], who compared patients treated with antibiotics during operation alone with those treated up to post-operative day (POD) 5, found that extended post-operative antibiotic prophylaxis was effective for the prevention of infectious complications. Togo et al. [16], in an RCT after liver resection, reported that two days were similarly effective compared with five days of antibiotic administration. In contrast, two other RCTs compared patients with and without post-operative antibiotics and found that postoperative antibiotic prophylaxis did not prevent post-operative infections after liver resection [17,18]. Many surgeons, however, tend to extend the duration of prophylactic antibiotic use and it seems that three or four days of antibiotic use is ordinary in Japan. Thus, the necessity of post-operative antibiotic prophylaxis or its duration after liver resection is not clear.

We introduced routine antibiotic prophylaxis for patients undergoing liver resection in 2006 with continued administration until POD 3. Beginning in March 2014, post-operative antibiotic agents were administered only once three to four hours after completion of the surgical procedure according to short-term use recommendations. In this study, we performed a retrospective analysis of the adequacy of the duration of post-operative antibiotic use and of the risk factors for infectious complications in consecutive patients undergoing liver resection without biliary reconstruction.

Patients and Methods

Patient characteristics and data collection

A total of 212 elective liver resections were performed between January 2011 and December 2015 at the Department of Digestive, Breast and Thyroid Surgery, Kagoshima University Hospital, of which four were performed with reconstruction of the bile duct. It seems that biliary reconstruction is a risk factor for post-operative bile leakage and infectious complication. Therefore, patients with biliary reconstruction were excluded in this study. Emergency operation with liver resection was not performed during the period of this study.

There were 115 patients between January 2011 and February 2014 who were given post-operative prophylactic antibiotic agents for three days after the operation (the POD 3 group). There were 93 patients between March 2014 and December 2015 who were given post-operative antibiotic agents only once three to four hours after the completion of operation (the POD 0 group). During the span of this study, the clinical pathway for treating patients undergoing liver resection was not changed except for the schedule of post-operative antibiotic administration.

To evaluate the influence of the duration of prophylactic antibiotic use on infectious complications, data from 115 patients in the POD 3 group and 93 patients in the POD 0 group were analyzed retrospectively. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Informed consent of therapeutic protocols, including operation method and post-operative management and use of clinicopathologic data for this cohort study were obtained from all patients.

A history of the presenting illness was obtained from all patients and a complete physical examination was performed. A self-administered questionnaire was used to gain information on medical history. Excessive alcohol consumption was defined as an average daily consumption of an amount equivalent to 70 g of pure ethanol over a period of more than 10 years. Diabetes mellitus diagnosis included patients whose blood sugar levels were controlled with medication. The American Society of Anesthesiologists (ASA) score was determined by an attending physician or an anesthesiologist.

Patient infectious status for hepatitis B and C viruses was determined by testing for the presence of the hepatitis B surface antigen and hepatitis C antibodies. Liver function was assessed by a liver biochemistry test, Child-Pugh grade, indocyanine green retention test (ICGR15), and technetium-99m-galactosyl human serum albumin scintigraphy [19]. Tumor progression was assessed by imaging studies, such as contrast enhanced computed tomography scans, magnetic resonance imaging, and ultrasonography. The resectability and extent of hepatic resection were evaluated according to the disease progression, liver function, and the general condition of the patients [20].

Protocol for antibiotic prophylaxis

Our clinical pathway for treating patients undergoing liver resection was changed according to the recommendations of the CDC guidelines as shown in our previous report [21]. From February 2006, for routine antibiotic prophylaxis, cefotiam 1.0 g, a second-generation cephalosporin, was administered 30 minutes before skin incision. In patients who underwent operations lasting longer than three hours, additional doses were given every three hours thereafter during the operation. This agent was also administered every 12 hours for three days post-operatively. After March 2014, we changed the protocol of post-operative administration, and the antibiotic agent was injected only once three to four hours after the completion of the operation according to the recommendations of short-term post-operative use [8,12].

All patients in this study underwent mechanical bowel preparation with magnesium citrate at 16.00 hours and sennoside at bedtime on the day before surgery. Oral antibiotic agents for bowel preparation were not used.

Surgical technique

The teams of surgeons were not changed during the period of this study. Liver resection was performed under intra-operative ultrasonographic guidance according to standard techniques. Anatomic resection was performed whenever possible, whereas partial resection was performed in consideration of limited liver functional reserve, anatomic location of the tumor, and benign disease. For hepatic parenchymal transection, we used an ultrasonic dissector (CUSA Excel^®; Integra Lifesciences Corporation, Plainsboro, NJ) to identify intra-hepatic structures such as hepatic veins and Glisson sheaths. Exposed structures of 2 mm or less were dissected using EnSeal^® (Ethicon Endo-Surgery, Cincinnati, OH) and larger vessels were ligated with 3-0 braided silk or vessel clips.

The Pringle maneuver (hepatic inflow occlusion time 15 minutes and reperfusion time five minutes) was performed during parenchymal transection. A bile-leak test, using a biliary tube inserted through the cystic duct, was routinely performed in patients in whom cholecystectomy was simultaneous. With this procedure, we identified small bile leakage sites on the cut liver surface and could repair these sites by suturing using 4-0 or 5-0 polydioxanone (PDS^® II) (Johnson & Johnson Corp., Tokyo, Japan). A 6Fr trans-cystic duct tube (C-tube, Sumius, Sumitomo Bakelite Co., Tokyo, Japan) was placed in case of wide cut parenchymal surfaces (e.g., central bisegmentectomy or right anterior sectionectomy) or with massive liver resection (e.g., trisectionectomy or extended hemi-hepatectomy). One or two closed drains (J-VAC, 7 mm flat, Johnson & Johnson Corp., Tokyo, Japan) were inserted at the end of the operation in the right sub-phrenic space or elsewhere close to the cut surface of liver parenchyma.

After closure of the fascia with interrupted sutures using 0 PDS II (Johnson & Johnson Corp., Tokyo, Japan), the surgical incisions in all patients were irrigated with 500 mL saline. The skin was closed using an interrupted sub-cuticular suture of 4-0 PDS II. According to the Couinaud classification [22] of liver segments, resection of three or more segments was considered a major resection, and resection of two or fewer segments was considered a minor resection.

Postoperative management

Normothermia was kept during an operation and after operation. Adequate control of serum blood glucose levels and peri-operative oxygenation (arterial blood oxygen saturation was at least 95 mm Hg) were also maintained.

Patients were allowed to drink on POD 1 and take food orally on POD 2. The management of central lines and urinary catheters was not changed during the period of this study—i.e., the urinary catheter was routinely removed on POD 2 or 3, and the central line was usually removed on POD 3 or 4. The intra-abdominal drains were removed on POD 2 or 3 after confirming hemostasis and the absence of bile leakage. The analysis of total bilirubin concentration in the drain fluid was routinely performed on POD 2. If the drain fluid was contaminated or bile stained, the drainage tube was retained until the bile leakage stopped or the intra-abdominal infection resolved. Post-operative bile leakage was defined according to the International Study Group of Liver Surgery published in 2011 [23].

Computed tomography was performed routinely to assess the remnant liver and infectious complications (e.g., intra-abdominal abscess or pneumonia) on POD 6 or 7. The C-tube was removed on POD 7 or 8 after confirming the lack of bile leakage.

Definition and diagnosis of infectious complications

Infectious complications were classified as combined SSI and remote site infection (RSI). SSI was defined as a condition in which purulent discharge was observed from any incision or space that was manipulated during an operation, within 30 days of that operation, with or without microbiologic evidence, according to the guideline issued by the CDC [8]. A RSI was defined as a condition in which fever and leukocytosis were present with bacteria in sputum, urine, catheter tip, blood, or bile, or according to the physician's judgment regardless of microbiologic evidence. Complications were graded in severity according to the original classification established by Dindo and Clavien [24,25].

Statistical analysis

Parametric data are presented as mean ± standard deviation and were compared using the unpaired t-test. Statistical analysis of non-parametric variables was performed using the two-sided or one-sided Fisher exact probability test, where appropriate. Risk factor analysis was performed using the logistic regression test. The cutoff values for continuous data in uni-variable and multi-variable analysis were defined as median values. Only significant variables in uni-variable analysis were included in the multivariable analysis. A p value of less than 0.05 was considered to be significant.

Results

Pre-operative characteristics and operative variables

The pre-operative characteristics of both groups are shown in Table 1. There were no significant differences in the pre-operative characteristics between the two groups. Table 2 shows operative characteristics and tumor factors. There were no significant differences in these characteristics, except for patients with combined resection of other organs without the gallbladder. Four patients in the POD 0 group simultaneously underwent resection of other organs: Partial resection of lung (n = 2), left nephrectomy (n = 1), distal pancreatectomy and splenectomy (n = 1). No patients in the POD 3 group underwent simultaneous resection of other organs.

Table 1.

Preoperative Characteristics of Patients

Variables	POD 0 (n = 93)	POD 3 (n = 115)	p
Age (years)	66.6 ± 10.6	65.5 ± 12.9	0.514
Male/Female	6330	77/38	>0.9999
BMI (kg/m²)	23.9 ± 3.7	23.9 ± 3.7	0.980
Diabetes mellitus, yes/no	23/70	3/78	0.282
Excessive alcohol consumption, yes/no	19 74	23/92	>0.9999
ASA score, 1/2/3	24/57/12	15/86/14	0.054
Liver diseases, HCC/ICC/metastasis/others	41/10/34/8	60/11/32/12	0.596
Pre-operative anti-cancer chemotherapy, yes/no	21/72	18/97	0.216
HBsAg, positive/negative	5/88	13/102	0.145
Anti-HCV-Ab, positive/negative	11/82	17/98	0.683
Child-Pugh classification, A/B	92 1	112/3	0.630
ICGR15 (%)	12.0 ± 5.2	12.6 ± 7.1	0.570
LHL15	0.915 ± 0.036	0.917 ± 0.031	0.677
Prothrombin time (%)	101.5 ± 11.5	99.1 ± 10.5	0.115
Albumin (g/dL)	4.1 ± 0.4	4.0 ± 0.5	0.195
Total bilirubin (mg/dL)	0.8 ± 0.5	0.8 ± 0.3	0.268
Platelets ( × 10⁴/μL)	17.9 ± 6.1	18.5 ± 6.4	0.473
AST (IU/L)	30.5 ± 15.3	33.1 ± 15.8	0.233
ALT (IU/L)	27.3 ± 17.9	31.3 ± 19.0	0.121
CRP	0.3 ± 0.8	0.3 ± 0.7	0.770

Values are mean ± standard deviation for parametric variables.

BMI = body mass index; ASA = American Society of Anesthesiologyists; HCC = hepatocellular carcinoma; ICC = intra-hepatic cholangiocarcinoma; HBsAg = hepatitis B surface antigen; HCV = hepatitis C virus; Ab = antibody; ICGR15 = indocyanine green retention rate at 15 min; LHL15 = liver uptake ratio by Tc GSA liver scintigraphy at 15 min; AST = aspartate aminotransferase; ALT = alanine aminotransferase; CRP = C-reactive protein; IU = internationall units.

Table 2.

Operative Variables and Tumor Factors of Patients

Variables	POD 0 (n = 93)	POD 3 (n = 115)	p
Primary hepatectomy, yes/no	86/7	111/4	0.225
Incision, laparotomy/laparoscopic	75/8	80/35	0.079
Operation methods			0.437
Extended hemi-hepatectomy	9	10
Hemi-hepatectomy	21	38
Sectionectomy	32	29
Segmentectomy	9	9
Partial resection	22	29
Operation types
Anatomic/non-anatomic	70/23	87/28	>0.9999
Major/minor	36/57	52/63	0.398
Combined resection of other organs except gallbladder, yes/no	4/89	0/115	0.039
Use of C-tube drainage, yes/no	30/63	34/81	0.763
Operative time (min)	404 ± 136	379 ± 127	0.186
Intra-operative blood loss (mL)	898 ± 896	1022 ± 898	0.323
Peri-operative transfusion, yes/no	39/54	54/61	0.486
Tumor size (cm)	3.9 ± 3.0	4.2 ± 3.4	0.426
Tumor number, simple/multiple	65/26	74/39	0.450
Pathological cirrhosis, yes / no	6/87	11/04	0.457

C-tube = transcystic duct tube.

Post-operative complications

Table 3 shows the post-operative complications and outcome. The SSIs were observed in 7.2% (15/208) of patients in this study—8% in the POD 0 group and 7% in the POD 3 group. Incisional SSIs, including superficial and deep SSIs, were observed in 5% of the POD 0 group and 3% of the POD 3 group (p = 0.517). Organ/space SSIs were observed in 2% of the POD 0 group and 3% of the POD 3 group (p = 0.694). The incidence of RSIs was 6% in the POD 0 group and 5% in the POD 3 group (p > 0.9999). Overall infectious complications, including both SSIs and RSIs, were observed in 12% of the POD 0 group and 11% of the POD 3 group (p > 0.9999).

Table 3.

Post-Operative Complications and Outcome

	POD 0 (n = 93)					POD 3 (n = 115)
		Clavien-Dindo grade					Clavien-Dindo grade
Variables	No. (%)	I	II	IIIa	IIIb	No. (%)	I	II	IIIa	IIIb	p
Infectious complications	11 (12)					13 (11)					>0.9999
Overall SSI	7 (8)		3	4		8 (7)		4	4		>0.9999
Incisional SSI	5 (5)		2	3		4 (3)		1	3		0.517
Organ/space SSI	2 (2)		1	1		4 (3)		3	1		0.694
Remote site infections	6 (6)					6 (5)					>0.9999
Pneumonia	2 (2)		2			1 (1)		1			0.588
Urinary tract infections	1 (1)		1			1 (1)		1			>0.9999
Catheter-related infections	0 (0)					2 (2)		2			0.503
Phlegmon of lower limbs	1 (1)		1			0 (0)					0.447
Biliary infections	2 (2)		2			2 (2)		2			>.9999
Bile leakage	3 (3)			2	1	3 (3)		1	1	1	>.9999
Pleural effusion	4 (4)			4		14 (12)	1	7	6		0.050
Ileus	2 (2)		1	1		2 (2)		2			>.9999
Intra-abdominal hemorrhage	1 (1)		1			3 (3)	1	2			0.630
Hyperbilirubinemia	1 (1)	1				1 (1)		1			>.9999
Portal vein thrombosis	1 (1)		1			1 (1)		1			>0.9999
Delirium	3 (3)	1	2			3 (3)	3				>0.9999
Re-laparotomy	1 (1)					1 (1)					>0.9999
Hospital death	0 (0)					0 (0)					>0.9999
Post-operative hospital stay (days)	13 ± 10					14 ± 9					0.503

Data of Clavien-Dindo grade are expressed as numbers of patients.

POD = post-operative day; SSI = surgical site infection.

Post-operative bile leakages were observed in three patients in each group. One patient with bile leakage in the POD 3 group underwent resection of the Spiegel lobe for intractable bile leakage at 48 days after initial right hemi-hepatectomy. One patient with bile leakage in the POD 0 group underwent re-operation for T-tube drainage because of post-operative perforation of the common bile duct at 11 days after initial extended left hemi-hepatectomy.

SSIs were observed in half (3/6) of the patients with post-operative bile leakage and half (2/4) of the patients with post-operative intra-abdominal hemorrhage (data not shown). The incidence of patients with post-operative pleural effusion necessitating diuretic medication (Clavien-Dindo grade II) or intervention without general anesthesia (Clavien-Dindo grade IIIa) tended to be higher in the POD 3 group than in the POD 0 group (p = 0.050). No re-operation was required except for the two patients with post-operative bile leakage. There was no hospital death and no significant difference in the post-operative hospital stay between the two groups.

Identified organisms causing infectious complications

Table 4 shows the causative organisms cultured from patients with post-operative infectious complications. Thirteen species of organisms were detected in 67% (15/24) of patients with infectious complications. The bacterial culture was not performed in five of 24 patients with infectious complications. Staphylococcus aureus was the most frequently cultured organism (three in the POD 0 group, four in the POD 3 group), with five being methicillin-resistant (one in the POD 0 group, four in the POD 3 group). There was no significant difference in the proportions of gram-positive or gram-negative organisms between the two groups. The proportion of organisms from intestinal flora and those from skin flora were similar in both groups.

Table 4.

Organisms Causing Infectious Complications

Organism	POD 0	POD 3
Gram-positive	6	10
MRSA^a	1	4
MSSA^a	2	0
Staphylococcus haemolyticus ^a	0	1
Enterococcus aerogenes ^b	0	1
Enterococcus faecalis ^b	1	0
Enterococcus faecium ^b	0	1
Streptococcus species ^b	1	1
Corynebacterium species ^a	1	1
Candida albicans	0	1
Gram-negative	3	2
Escherichia coli ^b	2	0
Klebsiella pneumoniae ^b	0	1
Enterobacter gergoviae ^b	0	1
Bacteroides fragilis ^b	1	0

Organisms belonged to skin flora.

Organisms belonged to intestinal flora.

Data are expressed as numbers. Duplications are included.

POD = post-operative day; MRSA = methicillin-resistant Staphylococcus aureus; MSSA = methicillin-sensitive S. aureus.

Risk factors for infectious complications, SSIs and bile leakage in multivariable analysis

Table 5 shows risk factors for each complication after multi-variable analysis. To evaluate the influence of the duration of prophylactic antibiotic use on infectious complications, risk factors for overall infectious complications, overall SSIs, incisional SSIs, and organ/space SSIs were analyzed. Operative time of ≥400 minutes was an independent risk factor for the development of infectious complications after liver resection. Repeat hepatectomy was an independent risk factor for incisional SSIs, and post-operative bile leakage and hemorrhage were independent risk factors for organ/space SSI development. For overall SSI development, age ≥68 years, bile leakage, and hemorrhage were independent risk factors.

Table 5.

Risk Factors for Infectious Complications, Surgical Site Infections and Bile Leakage in Multi-variable Analysis

Variables	Odds ratio	95% confidence intervals	p
Infectious complications
Age ≥68 years	1.97	0.78–4.97	0.152
Anatomic resection	4.46	0.54–36.58	0.164
Operative time ≥400 min	2.92	1.01–8.48	0.049
Overall SSI
Age ≥68 years	7.90	1.58–39.37	0.012
Bile leakage	17.26	1.22–58.0	0.031
Post-operative hemorrhage	8.41	1.18–252.0	0.037
Incisional SSI
Age ≥68 years	3.85	0.74–20.0	0.109
CRP ≥ 0.2	3.36	0.77–14.71	0.298
Repeat hepatectomy	9.17	1.39–62.50	0.021
Organ/space SSI
Age ≥68 years	1.16	0.98–1.37	0.077
Bile leakage	33.34	3.3–250	0.001
Post-operative hemorrhage	26.31	1.75–500	0.004
Bile leakage
ICC	5.41	0.92–31.88	0.062
CRP ≥ 0.2	5.42	0.56–52.13	0.144
Operative time ≥400 min	1.01	0.99–1.02	0.205

Only significant variables in uni-variable analysis were included in the multivariable analysis.

SSI = surgical site infection; ICC = intrahepatic cholangiocarcinoma; CRP = C-reactive protein.

Short-term post-operative antibiotic use (the POD 0 group) was not associated with development of infectious complications including both SSIs and RSIs. The odds ratio for the development of SSIs was high for bile leakage. Therefore, the analysis of bile leakage as a risk factor for post-operative SSIs was added to the multi-variable analysis. Although intra-hepatic cholangiocarcinoma, a C-reactive protein ≥0.2 mg/L, and longer duration of operation were significant risk factors for the occurrence of bile leakage in uni-variable analysis, no factor remained a significant independent risk factor for post-operative bile leakage in multi-variable analysis.

Discussion

The prevention of post-operative infections after liver resection is obviously important in peri-operative management, and it has been believed that antibiotic prophylaxis plays an important role. Antibiotic prophylaxis has been recommended by the CDC in its 1999 guideline [8]. In subsequent studies, administration within 60 minutes before skin incision has been generally recommended, with additional doses administered based on the antimicrobial half-life of the antibiotic agent or the amount of blood loss and fluids administered during longer operations [9,10].

Only one RCT on the efficacy of pre-operative antibiotic prophylaxis in elective liver resection has been performed [26]. Zhou et al. [26] compared patients who were treated with cefuroxime or physiologic saline (10 mL) within 30 minutes before skin incision and reported that pre-operative antibiotic prophylaxis resulted in no statistically significant benefit for post-operative infections including both SSIs and RSIs. In that study, however, antibiotics were only administered once in the prophylaxis group, and the incidence of overall infection (23.3% vs. 20.0%) and SSIs (13.3% vs. 15%) in both groups was relatively high. Therefore, it seems that pre-operative and intra-operative antibiotics prophylaxis has been useful to reduce post-operative complications in various different operations and thus we adopted the recommendation of the CDC.

To the best of our knowledge, thus far, four RCTs on the efficacy of post-operative antibiotic prophylaxis after liver resection have been conducted [15 –18]. Wu et al. [17] performed a comparative study of groups given and not given seven days of cefazolin and gentamicin. This study was reported in 1998 ahead of the CDC guideline. All patients in this study were treated with oral antibiotic agents for bowel preparation on the day before surgery and without pre-operative or intra-operative antibiotic prophylaxis. The infectious complication rate was 23% in both groups, and the authors found that post-operative antibiotic prophylaxis could not prevent post-operative infections after liver resection.

Sano et al. [15] compared groups treated with cefazolin during the surgical procedure alone, just before the skin incision and after the parenchymal transection, with a group also treated until POD 5 and found that post-operative antibiotic prophylaxis was effective for the prevention of post-operative signs of infection. In two other recent RCTs [16,18], pre-operative and intra-operative antibiotics were administered 30 minutes before operation, and additional agents were administered every three hours during the operation according to the recent guideline.

Togo et al. [16] compared patients treated with flomoxef sodium until POD 1 or POD 4 and reported that two days (operative day and POD 1) of antibiotic administration were sufficient for antibiotic prophylaxis after liver resection. In contrast, Hirokawa et al. [18] compared groups of patients who were treated with flomoxef sodium during operation alone and those treated until POD 3 and found no significant difference with respect to infectious complications, both SSIs, and RSIs. Thus, although the necessity for post-operative antibiotic prophylaxis or the duration of antibiotic administration after liver resection is contradictory, it seems that post-operative antibiotic prophylaxis is useful after digestive surgical procedures classified as clean-contaminated procedures, and we adopted the recommendation that antibiotic administration be stopped within 24 hours after operation [11,12].

Since we introduced routine antibiotic prophylaxis in liver resection, post-operative antibiotics were administered until POD 3. We changed the protocol for post-operative administration according to the recommendations of short-term postoperative use [8,12] after March 2014, however, and post-operative antibiotics were administered only once here to four hours after the completion of surgery to prevent the appearance of resistant strains. If an operation is prolonged or if blood loss is high, the tissue concentration of prophylactic antibiotic agents can be expected to be lower than the minimum inhibitory concentration (MIC). Intra-operative blood loss in liver resection is large compared with other digestive surgical procedures; therefore, we have added one dose of post-operative antibiotic agents to intra-operative administration to support the MIC until several hours after operation.

In this study, incisional SSIs were observed in 5% of the POD 0 group and 3% of the POD 3 group (p = 0.517), and organ/space SSIs were observed in 2% of the POD 0 group and 3% of the POD 3 group (p = 0.694). The incidence of SSIs was not higher than the rates of 3.0%–12.8% recently reported in other series [4,27 –29]. There was also no significant difference in the incidence of RSIs or overall infectious complications between the two groups. Thus, discontinuation of prophylactic antibiotic agents on the operative day appears to have no important influence on the incidence of infectious complications after liver resection.

Hirokawa et al. [18] reported that post-operative antibiotic prophylaxis after liver resection is unnecessary and costly according to the results of their RCT on the efficacy of post-operative antibiotic administration. We have added a single post-operative antibiotic dose, however, three to four hours after the completion of the operation. It seems that many surgeons conventionally tend to extend the duration of prophylactic antibiotic therapy making discontinuation of post-operative administration subjectively more difficult.

In several analyses of patients without post-operative prophylactic antibiotic agents after liver resection, the incidence of incisional SSIs was reported to be 3.2%–12.8% and that of organ/space SSIs was reported to be 1.0%–4.3% [18,29,30]. The incidence of incisional and organ/space SSIs in this study was not higher in patients without post-operative antibiotic administration. Further studies of the necessity of post-operative antibiotic prophylaxis are needed.

Cefotiam is a second-generation cephalosporin antibiotic that has a broad spectrum including both gram-positive and gram-negative bacteria with low toxicity and is widely used as a prophylactic antibiotic agent in Japan. Although it has a strong antibacterial effect against staphylococcal bacteria, it lacks an antibiotic effect against Pseudomonas aeruginosa and Bacteroides fragilis [31].

In this study, 13 species of organism were detected in patients with infectious complications, and there were no significant differences between the two groups in the proportion of gram-positive versus gram-negative organisms, or in the proportion of intestinal versus skin flora. Methicillin-resistant S. aureus was detected more often in the POD 3 group than in the POD 0 group. The purpose of shorter duration of prophylactic antibiotic administration is to minimize the evolution of antibiotic-resistant bacteria; discontinuation of post-operative prophylactic antibiotics on the operative day should be recommended.

In multivariable analyses, the short-term post-operative antibiotic use (the POD 0 group) was not associated with development of infectious complications including both SSIs and RSIs. Repeat hepatectomy was an independent risk factor for incisional SSIs, and post-operative bile leakage and hemorrhage were independent risk factors for organ/space SSIs. For overall development of infectious complications including both SSIs and RSIs, an operative duration >400 minutes remained as an independent risk factor.

Eleven of the 208 patients in this study underwent repeat hepatectomy. Incisional SSIs were found in 18% of the 11 patients having a repeat hepatectomy and 3.5% of patients having a primary hepatectomy. Sadamori et al. [4] reported that repeat hepatectomy was a risk factor for organ/space SSIs, and it is important to establish treatment strategies for reducing the rate of SSIs in the case of repeat hepatectomy.

Our previous study [21] and several other studies [4,32 –34] have reported that post-operative bile leakage after liver resection was the key independent risk factors for SSIs. In recent studies in Japan, the incidence of post-operative bile leakage after liver resection was reported to be 4.8%–12.8% [4,27,28,34,35]. In our experience in this study, the incidence of bile leakage was 2.9% (6/208), which is consistent with reports from other high volume centers. Multi-variable analysis, however, revealed that post-operative bile leakage was the strongest risk factor for SSIs. The presence of bile in the dead space after hepatic resection may constitute an ideal environment for bacterial growth and the development organ/space SSIs. Specifically, in this study, all SSIs in cases of bile leakage were organ/space SSIs.

Several studies on the efficacy of the C-tube for reducing the incidence of post-operative bile leakage after liver resection have been reported [36,37] and we have aggressively placed C-tubes in cases with a large surface area of cut parenchyma or with massive liver resection. Although the usefulness of he C-tube in preventing post-operative bile leakage and SSIs could not be confirmed in this study, C-tube drainage after liver resection might have contributed to a low incidence of bile leakage in this study. It is crucial to prevent bile leakage to reduce post-operative SSIs. Further studies are needed.

The present study was associated with some limitations. First, although we changed the protocol of post-operative antibiotic administration for all patients who underwent liver resection after 2014 according to the recommendations of short-term post-operative use, this study was a retrospective, non-randomized, and observational study. The only significant difference between the two groups, however, was in the number of patients with simultaneous resection of other organs; there was no difference in their pre-operative backgrounds, operative variables, or tumor factors. The strength of this study is that all patients undergoing elective liver resection were enrolled; only patients with biliary reconstruction were eliminated.

In addition, cefotiam is not marketed in a part of Western countries. It is a medicine available in a number of Eastern countries and several European countries, however, and is widely used as a prophylactic antibiotic for surgical procedures of various digestive diseases in Japan.

Conclusion

Our study demonstrated that pre-operative, intra-operative, and immediate post-operative prophylactic administration of cefotiam was sufficient to optimize SSI incidence. No additional significant benefit was conferred by prolonging administration until post-operative day 3. Further studies of the necessity of post-operative antibiotic prophylaxis on the operative day are required to further reduce medical cost and the evolution of antibiotic-resistant bacteria.

Footnotes

Acknowledgments

We thank Koji Minami, Kiyokazu Hiwatashi, Motoyuki Hashiguchi, and Shinichiro Mori for their attendance in enrolling patients and collecting data.

Author Disclosure Statement

No competing financial interests exist.

References

Poon

, Fan

, Yu

, et al. A prospective longitudinal study of quality of life after resection of hepatocellular carcinoma. Arch Surg, 2001; 136:693–699.

Fan

, Lo

, Liu

, et al. Hepatectomy for hepatocellular carcinoma: Toward zero hospital deaths. Ann Surg, 1999; 229:322–330.

Torzilli

, Makuuchi

, Inoue

, et al. No-mortality liver resection for hepatocellular carcinoma in cirrhotic and noncirrhotic patients: Is there a way? A prospective analysis of our approach. Arch Surg, 1999; 134:984–992.

Sadamori

, Yagi

, Shinoura

, et al. Risk factors for major morbidity after liver resection for hepatocellular carcinoma. Br J Surg, 2013; 100:122–129.

Virani

, Michaelson

, Hutter

, et al. Morbidity and mortality after liver resection: Results of the patient safety in surgery study. J Am Coll Surg, 2007; 204:1284–1292.

McGowan

Jr . Cost and benefit of perioperative antimicrobial prophylaxis: Methods for economic analysis. Rev infect Dis, 1991; 13(Suppl):S879–S889.

Taylor

, Kirkland

, McKenzie

, et al. The effect of surgical wound infection on postoperative hospital stay. Can J Surg, 1995; 38:149–153.

Mangram

, Horan

, Pearson

, et al. Guideline for prevention of surgical site infection, 1999. Hospital Infection Control Practices Advisory Committee. Infect Control Hosp Epidemiol, 1999; 20:250–278.

Alexander

, Solomkin

, Edwards

. Updated recommendations for control of surgical site infections. Ann Surg, 2011; 253:1082–1093.

10.

Hawn

, Richman

, Vick

, et al. Timing of surgical antibiotic prophylaxis and the risk of surgical site infection. JAMA Surg, 2013; 148:649–657.

11.

Park

, Kim

, Park

, et al. Influence of shorter duration of prophylactic antibiotic use on the incidence of surgical site infection following colorectal cancer surgery. Ann Coloproctol, 2015; 31:235–242.

12.

Bratzler

, Dellinger

, Olsen

, et al. Clinical practice guidelines for antimicrobial prophylaxis in surgery. Am J Health Syst Pharm, 2013; 70:195–283.

13.

Baum

, Anish

, Chalmers

, et al. A survey of clinical trials of antibiotic prophylaxis in colon surgery: Evidence against further use of no-treatment controls. N Engl J Med, 1981; 305:795–799.

14.

Song

, Glenny

. Antibiotic prophylaxis in colorectal surgery: A systematic review of randomized controlled trials. Br J Surg, 1998; 85:1232–1241.

15.

Sano

, Takayama

, Makuuchi

. Prophylactic antibiotics in hepatectomy. Geka (Surgery), 2002; 64:1635–1639.

16.

Togo

, Tanaka

, Matsuo

, et al. Duration of antimicrobial prophylaxis in patients undergoing hepatectomy: A prospective randomized controlled trial using flomoxef. J Antimicrob Chemother, 2007; 59:964–970.

17.

, Yeh

, Lin

, et al. Prospective randomized trial of systemic antibiotics in patients undergoing liver resection. Br J Surg, 1998; 85:489–493.

18.

Hirokawa

, Hayashi

, Miyamoto

, et al. Evaluation of postoperative antibiotic prophylaxis after liver resection: A randomized controlled trial. Am J Surg, 2013; 206:8–15.

19.

Iguchi

, Sato

, Kouno

, et al. Comparison of Tc-99m-GSA scintigraphy with hepatic fibrosis and regeneration in patients with hepatectomy. Ann Nucl Med, 2003; 17:227–233.

20.

Makuuchi

, Kosuge

, Takayama

, et al. Surgery for small liver cancers. Semin Surg Oncol, 1993; 9:298–304.

21.

Sakoda

, Ueno

, Iino

, et al. Risk factors associated with surgical site infection after hepatectomy for liver disease. J US-China Med Sci, 2014; 11:1–8.

22.

Couinaud

Surgical Anatomy of the Liver Revised. Paris: Acheve D'imprimer Sur Les Presses, 1989.

23.

Koch

, Garden

, Padbury

, et al. Bile leakage after hepatobiliary and pancreatic surgery: A definition and grading of severity by the International Study Group of Liver Surgery. Surgery, 2011; 149:680–688.

24.

Dindo

, Demartines

, Clavien

. Classification of surgical complications: A new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg, 2004; 240:205–213.

25.

Clavien

, Barkun

, de Oliveira

, et al. The Clavien-Dindo classification of surgical complications: Five-year experience. Ann Surg, 2009; 250:187–196.

26.

Zhou

, Chen

, Li

, et al. Preoperative antibiotic prophylaxis does not reduce the risk of postoperative infectious complications in patients undergoing elective hepatectomy. Dig Dis Sci, 2016; 61:1707–1713.

27.

Tsujita

, Yamashita

, Takeishi

, et al. Subcuticular absorbable suture with subcutaneous drainage system prevents incisional SSI after hepatectomy for hepatocellular carcinoma. World J Surg, 2012; 36:1651–1656.

28.

Nanashima

, Arai

, Oyama

, et al. Associated factors with surgical site infections after hepatectomy: Predictions and countermeasures by a retrospective cohort study. Int J Surg, 2014; 12:310–314.

29.

Kokudo

, Uldry

, Demartines

, Halkic

. Risk factors for incisional and organ space surgical site infections after liver resection are different. World J Surg, 2015; 39:1185–1192.

30.

Kobayashi

, Gotohda

, Nakagohri

, et al. Risk factors of surgical site infection after hepatectomy for liver cancers. World J Surg, 2009; 33:312–317.

31.

Kusachi

, Sumiyama

, Nagao

, et al. Prophylactic antibiotics given within 24 hours of surgery, compared with antibiotics given for 72 hours perioperatively, increased the rate of methicillin-resistant Staphylococcus aureus isolated from surgical site infections. J Infect Chemother, 2008; 14:44–50.

32.

Okabayashi

, Nishimori

, Yamashita

, et al. Risk factors and predictors for surgical site infection after hepatic resection. J Hosp Infect, 2009; 73:47–53.

33.

Togo

, Kubota

, Takahashi

, et al. Usefulness of absorbable sutures in preventing surgical site infection in hepatectomy. J Gastrointest Surg, 2008; 12:1041–1046.

34.

Arikawa

, Kurokawa

, Ohwa

, et al. Risk factors for surgical site infection after hepatectomy for hepatocellular carcinoma. Hepatogastroenterology, 2011; 58:143–146.

35.

Harimoto

, Shirabe

, Abe

, et al. Prospective randomized controlled trial investigating the type of sutures used during hepatectomy. World J Gastroenterol, 2011; 17:2338–2342.

36.

Hotta

, Kobayashi

, Taniguchi

, et al. Postoperative evaluation of C tube drainage after hepatectomy. Hepatogastroenterology, 2003; 50:485–490.

37.

Nanashima

, Abo

, Shibuya

, et al. Does the placement of a cystic duct tube after a hepatic resection help reduce the incidence of post-operative bile leak?. HPB (Oxford), 2013; 15:517–522.