Comparison of Concentrations of Sulbactam-Ampicillin Administered by Bolus Injections or Bolus plus Continuous Infusion in Tissues of Patients Undergoing Colorectal Surgery

Abstract

The concentrations of sulbactam and ampicillin were determined in sera and different abdominal tissues of 16 patients who underwent elective colorectal surgery. Patients were randomly allocated to two groups. At the time of induction of anesthesia, patients in group 1 (eight patients) were given 1,000 mg of sulbactam with 2,000 mg of ampicillin by intravenous bolus injection (3 min). This dose was administered again after 2 h by bolus injection by the same route. Patients in group 2 (eight patients) were given the same initial dose of sulbactam-ampicillin by bolus injection (3 min). Then, a continuous infusion of 1,000 mg of sulbactam with 2,000 mg of ampicillin in normal saline was immediately started and was administered over a 4-h period. Blood samples were collected to determine peak (10 min) and trough (end of surgery) antibiotic levels. Serial blood samples were also collected at predetermined periods (at the time of opening and closing of the abdominal cavity and at the time of surgical anastomosis). Abdominal wall fat, epiploic fat, and colonic wall tissue samples were collected simultaneously. Antibiotic concentrations were determined by high-performance liquid chromatography. Similar levels of the drugs in serum were observed for the two regimens of administration, with trough sulbactam levels of 33 ± 16 and 37 ± 22 μg/ml in groups 1 and 2, respectively, and trough ampicillin levels of 72 ± 55 and 79 ± 47 μg/ml in groups 1 and 2, respectively. Similar sulbactam concentrations were observed in abdominal tissues whichever regimen of administration was used; in fatty tissues the sulbactam concentrations ranged from 2.7 to 3.8 μg/g for group 1 and from 1.7 to 4.0 μg/g for group 2, and sulbactam concentrations in the colonic wall were 5.6 ± 7.7 and 6.8 ± 3.2 μg/g in groups 1 and 2, respectively (not significant). Again, no influence of the regimen of administration was observed on tissue ampicillin concentrations; in fatty tissues ampicillin concentrations ranged from 4.1 to 5.4 μg/g for group 1 and from 3.2 to 5.8 μg/g for group 2, and sulbactam concentrations in the colonic wall were 7.0 ± 2.8 and 11.0 ± 4.7 μg/g for groups 1 and 2, respectively (not significant). In most patients, the concentrations of ampicillin-sulbactam were greater than the MIC at which 50% of isolates are inhibited (MIC₅₀) for Bacteroides fragilis in the fatty tissues. In the colonic wall, for most patients the concentrations of ampicillin-sulbactam were greater than the MIC₉₀ for B. fragilis. No influence of the regimen of administration was observed on the ratio of the two components in the tissues investigated and in sera. In conclusion, a second intraoperative bolus injection or a continuous infusion were equally effective in maintaining sulbactam-ampicillin concentrations in abdominal tissues. The first method of administration can be recommended since it is easier to handle.

Colorectal surgery is a contaminated aseptic type of surgery, and prophylactic antibiotics are widely used prior to and during surgery. The importance of the prophylactic antibiotics in reducing postoperative infections and mortality has been demonstrated (4, 15, 22). Antibiotics are selected on the basis of their ability to eradicate the bacteria most likely to contaminate a surgical wound: Escherichia coli and various members of the family Enterobacteriaceae, Enterococcus faecalis, Staphylococcus aureus, Bacteroides fragilis, and other anaerobic bacteria (4, 15, 22). The basic concept underlying the optimal use of antibiotic prophylaxis is that an adequate amount of an appropriate antibiotic should be present in the blood and tissues before a bacterial challenge. It seems clear that during a surgical procedure, adequate antibiotic concentrations should be achieved at all sites of potential infection. Also, antibiotic activity should be maintained throughout the procedure, from incision to closure (7, 9, 18, 19). The combination of sulbactam-ampicillin can be selected for use as a prophylactic antibiotic in abdominal surgery (10, 11, 23). It is effective in the prevention of wound infections in operations in which the most likely pathogens are gram-negative, anaerobic, or mixed bacteria (20). However, sulbactam-ampicillin has a short half-life (1 h) (8, 12), and for longer procedures such as colorectal surgery, antibiotic concentrations may not be maintained until the end of surgery when the first bolus dose is administered at the time of induction of anesthesia (7, 18, 19). When intermittent antibiotic dosing is used, antibiotic levels in serum and tissues will gradually decrease and the bacteria will often be exposed to low concentrations during the dosing interval. The present study was conducted during colorectal surgery to compare the antibiotic activities of two different regimens of intraoperative antibiotic administration: the use of a second bolus injection 2 h after the beginning of surgery or the use of a continuous antibiotic infusion during surgery. We wanted to confirm the maintenance of antibacterial concentrations of the antibiotics at potential sites of infection.

MATERIALS AND METHODS

Patients and drugs.

This study received the approval of the ethical committee of our institution (Hopital Nord), and all patients gave their informed consent. Fifteen patients were scheduled for elective colorectal surgery. All patients had normal hepatic function (serum bilirubin level, <10 μmol/liter) and renal function (creatinine clearance, >100 ml/min/1.74 m²). None had a history of allergic reaction to β-lactam antibiotics. None presented with any clinical sign (normal body temperature, normal physical examination) or laboratory sign (normal leukocyte count) of infection or had received antibiotic treatment in the preceding 3 weeks. Patients were randomly assigned to one of two groups. Patients in group 1 (five males and three females; mean age, 57 ± 12 years; mean body weight, 65 ± 10 kg) were given 1,000 mg of sulbactam with 2,000 mg of ampicillin by intravenous bolus injection (3 min). This dose was administered again after 2 h. Patients in group 2 (six males and two females; mean age, 59 ± 14 years; mean body weight, 67 ± 11 kg) were given by intravenous bolus injection (3 min) the same initial dose of sulbactam-ampicillin used for group 1 patients, and a continuous infusion of 1,000 mg of sulbactam with 2,000 mg of ampicillin in normal saline was immediately started and administered over a 4-h period.

Specimen collection.

Blood samples (10 ml each) were collected from a central venous catheter before surgery, 10 min after the beginning of surgery (peak level), and at the end of surgery (trough level). Blood samples were kept on ice and were centrifuged in a refrigerated centrifuge within 30 min of collection. Sera were placed in polypropylene tubes. Several samples of the selected tissues were collected during surgery. Abdominal wall fat and epiploic fat were obtained at the times of opening and closure of the abdominal cavity; colonic wall and epiploic fat samples were obtained at the time of surgical anastomosis. The exact time of sample collection was registered, and a corresponding blood sample was taken. Tissue sample sizes were 1 cm³ or larger. To remove the attached blood, all tissue samples were rinsed in sterile saline and were cleaned with dry, sterile gauze. Tissue samples were placed in sterile polypropylene tubes. Serum and tissue samples were assayed immediately following surgery.

Specimen preparation procedures.

For determination of sulbactam levels in serum, 1 ml of the sample was mixed with 0.5 ml of imidazole (2.5 M; pH 6.8) (5, 21). The solution was stored for 24 h at 4°C. After mixing (with a Vortex mixer for 30 s) the sample was deproteinized by the addition of 2.0 ml of acetonitrile. After mixing (15 s) and centrifugation at 3,000 rpm (Sorvall centrifuge) for 10 min, acetonitrile was removed by extraction with 3.0 ml of dichloromethane. Then, the solution was vortexed (30 s) and centrifuged at 3,000 rpm (Sorvall centrifuge) for 5 min. The aqueous phase (50 μl) was injected onto the high-performance liquid chromatography (HPLC) system.

For determination of ampicillin levels in serum, 1 ml of the sample was mixed with 1 ml of acetonitrile. After mixing and centrifugation the sample was then treated in the same way that the sulbactam samples were. The aqueous phase (50 μl) was injected onto the HPLC system.

For determination of the sulbactam and ampicillin levels in the tissues, the samples were weighed and 500 mg was homogenized with an Ultra-Turrax homogenizer at 4.0°C for 45 s. One milliliter of distilled water was used for the tissue homogenization process. Centrifugation of the homogenate yielded a clear supernatant which was treated like serum samples were.

HPLC systems.

The sulbactam concentrations were determined by using a dual-column HPLC system with UV detection (5, 21). The precolumn (LiChrospher RP 18 E, 5 mm; 25 by 4 mm [inner diameter]) was connected to the analytical column (LiChrospher RP 18 E, 5 μm; 125 by 4 mm [inner diameter]). The mobile phase consisted of 2% acetonitrile and 98% 0.1 M potassium phosphate buffer (pH 6.1). The flow rate was 1.0 ml/min, and sulbactam was detected at 322 nm.

The ampicillin concentrations were determined by using the same HPLC systems described above. The mobile phase consisted of 5% acetonitrile and 95% 0.01 M potassium phosphate buffer (pH 4.7). The flow rate was 1.6 ml/min, and ampicillin was detected at 219 nm. The lower limit of detection of sulbactam and ampicillin was 0.1 μg/ml for tissue and serum samples. The percent recoveries for sulbactam were 89% ± 4% and 91% ± 3% for serum and tissue samples, respectively. The percent recoveries for ampicillin were 94% ± 2% and 93% ± 4% for serum and tissue samples, respectively. For sulbactam, the within-day and between-day (3 days) interassay precisions were 1.43 and 1.41% (20 μg/ml) and 2.79 and 2.77% (40 μg/ml) for serum samples, respectively, and 3.3 and 5.0% (3 μg/g) and 4.2 and 6.4% (0.5 μg/g) for tissue samples, respectively. For ampicillin the within-day and between-day (3 days) interassay precisions were 1.46 and 1.20% (25 μg/ml) and 1.77 and 1.24% (50 μg/ml) for serum samples, respectively, and 3.7 and 5.2% (15 μg/g) and 5.3 and 6.7% (1 μg/g) for tissue samples, respectively.

Evaluation of sulbactam:ampicillin ratios.

Samples of specific tissues were analyzed according to collection time: opening and closure of the abdominal cavity and surgical anastomosis of the colon. For each defined period, the ratios of the concentrations of the two compounds (sulbactam and ampicillin) were obtained for the different tissue samples and the corresponding serum samples. The mean ± the standard deviation ratio was calculated for both groups. Antibiotic concentrations in tissues were also compared to the MICs at which 50% (MIC₅₀s) and 90% (MIC₉₀s) of Bacteroides fragilis isolates are inhibited (2 and 8 mg/liter for sulbactam and ampicillin, respectively), assuming a 1:2 sulbactam-to-ampicillin ratio.

Clinical evaluation.

All patients were closely monitored during their hospital stays and for up to 30 days for clinical or laboratory signs of infection.

Statistical analysis.

Statistical analysis was performed by Student’s t test for unpaired values and the chi-square test. A P value of less than 0.05 was considered significant.

RESULTS

Concentrations in serum.

The sulbactam and ampicillin concentrations in serum are presented in Table 1. No significant differences in concentrations in serum were observed, whichever drug regimen used. At the end of surgery both methods of drug administration achieved high levels of sulbactam and ampicillin in the two groups.

TABLE 1.

Concentrations of sulbactam and ampicillin in serum^a

Concn	Group 1			Group 2
	Sulbactam concn (μg/ml)	Ampicillin concn (μg/ml)	Sulbactam concn/ampicillin concn ratio	Sulbactam concn (μg/ml)	Ampicillin concn (μg/ml)	Sulbactam concn/ampicillin concn ratio
Peak (10 min)	142 ± 9	267 ± 155	0.5 ± 0.1	109 ± 8	207 ± 85	0.6 ± 0.2
Trough (end of surgery)b	33 ± 16	72 ± 52	0.5 ± 0.2	37 ± 22	79 ± 47	0.5 ± 0.2

^aValues are means ± standard deviations. 

^bGroup 1, 184 ± 11 min; group 2, 175 ± 16 min. 

Antibiotic concentrations in tissues.

The sulbactam and ampicillin concentrations in tissues at various periods during surgery are presented in Table 2. Stable concentrations of antibiotics were observed in fatty tissues when the concentrations at the surgical opening and closure periods are compared. This was observed after administration of a second bolus dose or the use of the continuous infusion during surgery. No significant differences were observed between the two groups. Sulbactam and ampicillin concentrations were higher in the colonic wall than in fatty tissues. The mode of antibiotic administration during surgery had no influence on the concentrations achieved in colonic wall samples.

TABLE 2.

Concentrations of sulbactam and ampicillin in abdominal tissues and serum of patients undergoing colorectal surgery at different periods of the surgical procedure

Group and druga	Mean ± SD drug concn at the following times:
	Openingb			Surgical anastomosisc			Closured
	Abdominal wall fat (μg/g)	Epiploic fat (μg/g)	Serum (μg/ml)	Colonic wall (μg/g)	Epiploic fat (μg/g)	Serum (μg/ml)	Abdominal wall fat (μg/g)	Epiploic fat (μg/g)	Serum (μg/ml)
Group 1d
Sulbactam	3.1 ± 3.8	3.1 ± 2.7	47 ± 18	5.6 ± 7.7e	3.7 ± 6.8	34 ± 21	2.7 ± 2.5	3.0 ± 3.4	32 ± 14
Ampicillin	5.1 ± 2.5	5.3 ± 3.1	88 ± 29	7.0 ± 2.8e	4.9 ± 2.2	74 ± 36	5.4 ± 1.3	4.1 ± 2.7	75 ± 50
Group 2d
Sulbactam	4.0 ± 4.2	3.4 ± 2.3	55 ± 23	6.8 ± 3.2e	2.8 ± 1.9	37 ± 12	1.7 ± 1.2	1.8 ± 1.2	38 ± 14
Ampicillin	5.8 ± 1.1	5.7 ± 3.3	120 ± 47	11.0 ± 4.7e	4.5 ± 2.0	83 ± 47	3.5 ± 2.0	3.2 ± 1.1	74 ± 49

^aBolus and infusion doses were 1,000 mg for sulbactam and 2,000 mg for ampicillin. 

^bFor groups 1 and 2, the opening times in relation to the start of antibiotic administration were 35 ± 6 and 36 ± 7 min, respectively. 

^cFor groups 1 and 2, the surgical anastomosis times in relation to the start of antibiotic administration were 124 ± 12 and 117 ± 13 min, respectively. 

^dFor groups 1 and 2, the closure times in relation to the start of antibiotic administration were 184 ± 11 and 175 ± 16 min, respectively. 

^eP < 0.02 compared with concentrations in fatty tissues. 

Sulbactam-to-ampicillin concentration ratios.

The sulbactam-to-ampicillin concentration ratios are presented in Tables 1 and 3. For group 1, they ranged from 0.33 to 0.70 in tissues and from 0.44 to 0.67 in serum. For group 2, the ratios ranged from 0.41 to 0.69 in tissues and from 0.46 to 0.67 in serum. The ratios were not significantly modified by the method of administration of antibiotics during surgery.

TABLE 3.

Sulbactam concentration/ampicillin concentration ratios at different periods of the surgical procedure in patients undergoing colorectal surgery

Groupa	Mean ± SD sulbactam concn/ampicillin concentration ratio at the following times:
	Opening			Surgical anastomosis			Closure
	Abdominal wall fat	Epiploic fat	Serum	Colonic wall	Epiploic fat	Serum	Abdominal wall fat	Epiploic fat	Serum
Group 1	1.2 ± 2.1	0.60 ± 0.60	0.52 ± 0.11	1.8 ± 3.7	1.3 ± 2.8	0.47 ± 0.09	0.6 ± 0.6	1.0 ± 1.5	0.44 ± 0.12
Group 2	0.80 ± 0.90	0.70 ± 0.70	0.67 ± 0.11	0.80 ± 0.60	0.80 ± 0.80	0.46 ± 0.11	0.60 ± 0.50	0.60 ± 0.50	0.52 ± 0.09

^aBolus and infusion doses were 1,000 mg for sulbactam and 2,000 mg for ampicillin. 

Table 4 presents data for the patients who achieved ampicillin-sulbactam concentrations in tissue greater than the MIC for B. fragilis. For most patients concentrations in the fatty tissues greater than the MIC₅₀ were achieved, but only in the colonic wall were concentrations greater than the MIC₉₀ for B. fragilis achieved.

TABLE 4.

Numbers of patients with concentrations in tissue greater than or equal to the MIC₅₀ and MIC₉₀ for B. fragilis at the different periods of the surgical procedure

Group	No. of patients with the indicated concn/total no. of patients
	Opening				Surgical anastomosis				Closure
	Abdominal wall fat		Epiploic fat		Colonic wall		Epiploic fat		Abdominal wall fat		Epiploic fat
	MIC50	MIC90	MIC50	MIC90	MIC50	MIC90	MIC50	MIC90	MIC50	MIC90	MIC50	MIC90
Group 1	8/8	1/8	8/8	1/8	8/8	5/8	8/8	1/8	7/8	1/8	8/8	0/8
Group 2	8/8	1/8	8/8	1/8	8/8	7/8	7/8	1/8	7/8	1/8	6/8	0/8

No patient had any signs of postoperative infection during the 30-day follow-up period.

DISCUSSION

This study shows that the intraoperative administration of sulbactam-ampicillin as two sequential bolus injections or by bolus injection plus a continuous infusion during surgery had no significant influence on antibiotic penetration into tissue. Previous studies have established the basic principles of antibiotic prophylaxis in surgical procedures, and the main points are that (i) the antibiotic must reach the tissues involved before surgery allows bacterial contamination, and (ii) the drug must attain and maintain concentrations in serum and tissues high enough to inhibit the growth of contaminating pathogens (7, 9, 15, 22). This study addressed the tissue penetration of a β-lactam–β-lactamase inhibitor combination, with the drugs being administered as two sequential bolus injections or one bolus injection immediately followed by a continuous infusion during the intraoperative period. The absolute concentrations of the drugs in tissues, the time course of these concentrations, and the extent of penetration are of interest. Penetration was evaluated by measuring the concentration in tissue-to-concentration in plasma ratios at specific collection times. The data for sulbactam and ampicillin are in agreement with results reported previously (11, 13, 14). Specific differences between tissues were observed. The extent of sulbactam and ampicillin penetration into fat was 4 to 8% of the levels in serum, significantly lower than the penetration into the colonic wall (10 to 20%; P < 0.02). This is in agreement with a study on the penetration of tazobactam-piperacillin into fat, which was 10% of the levels in plasma (16), but is at variance with a study with clavulanic acid-amoxicillin, in which the former drug had a higher level of penetration (30%) than the latter drug (10 to 20%) (17). The exact mechanisms that cause these important findings deserve further investigations. The level of penetration of sulbactam and ampicillin into the colonic wall was higher than that into fatty tissues. This might be explained by the fact that the blood flow to the gastrointestinal tract is higher than that to the fat. From these data, it is clear that no single mechanism governs the extent of penetration of antibiotics into tissues. Further investigations are needed to more clearly define the relationship among blood flow, water content, and the physicochemical properties of the agents.

In the present study, the extent of penetration of both compounds (sulbactam and ampicillin) was maintained throughout the surgical procedure thanks to the intraoperative administration of the antibiotics. In clinical practice, when intermittent antibiotic dosing is used, antibiotic levels in serum and tissues will gradually decrease. The bacteria will often be exposed to low concentrations of the antibiotic during the dosing interval and infection may start in the tissues potentially affected by postoperative infection. This emphasizes the need for repeated administration of drugs with shorter half lives (7, 15, 17, 22), such as sulbactam and ampicillin (8, 12). If the intraoperative administration is not performed, decreased concentrations of both compounds, sulbactam and ampicillin, would probably be measured in fatty tissues and colonic wall, with an increased risk of postoperative infection. To maintain active concentrations of antibiotics, one can consider either the administration of a second intraoperative bolus injection or continuous administration during surgery. The administration of an intraoperative bolus dose (after 2 h, which corresponds to two half-lives for most antibiotics used) is recommended, but objective data are lacking (7, 15, 22). β-Lactam antibiotics can also be administered by continuous infusion (6), and in the present study, both routes of administration were compared. With either route, the same concentrations in serum (including trough levels), the same antibiotic concentrations in tissues, the same concentration in tissue/concentration in serum ratios, and the same sulbactam-to-ampicillin concentration ratios were achieved. Thus, both administration regimens were equally effective at preventing fluctuating antibiotic concentrations during colorectal surgery. Since the administration of a second intraoperative dose is easier to handle than a continuous infusion, the former route of administration can be selected for clinical use.

Current practice with β-lactam–β-lactamase inhibitor combinations is to maintain an optimal ratio between the two components (1:2 sulbactam to ampicillin) (3). Because of their similar pharmacokinetics these two components theoretically maintain a 1:2 ratio at the site of potential infection over the entire period of surgery. In the present study, ratios close to 1:2 were obtained in the tissues studied, but it has been suggested that the efficacy of sulbactam-ampicillin is more dependent upon the maintenance of both sulbactam and ampicillin levels above a minimum critical concentration than on the maintenance of a constant 1:2 ratio (2). With both regimens of antibiotic administration used in the present study, the levels of sulbactam and ampicillin were maintained in the studied tissues until the end of surgery. Will these concentrations be active against the most common pathogens encountered during surgery? It can be speculated that a desirable goal is to achieve antibiotic levels above the MIC for the potential pathogen. This also means maintaining such levels until the end of the surgical procedure (7, 19). At the time of closure of the abdomen, sulbactam concentrations in the fatty tissues reached an average of 2 μg/g. At this concentration of sulbactam, the ampicillin MIC for plasmid-mediated β-lactamase-producing strains of E. coli can be as high as 16 μg/ml (1). Given the average level of ampicillin achieved in the same tissues (Table 2), some patients can be at risk of postoperative infection. This finding was observed whatever the regimen of antibiotic administration used. With regard to B. fragilis, in most patients, antibiotic levels greater than the MIC₅₀ were achieved in the fatty tissues up to the end of surgery, but only in the colonic wall were antibiotic concentrations greater than the MIC₉₀ for this organism achieved. From a clinical point of view, in the present study no patient exhibited any clinical or laboratory signs of postoperative infection during the 30-day follow-up period.

In conclusion, the results of this study indicate that the intraoperative administration of sulbactam-ampicillin is able to maintain constant antibiotic levels in tissue during surgery. After the administration of a first bolus injection 30 min prior to surgery, antibiotic levels can be maintained either by the injection of a second bolus dose after 2 h (two half-lives) or by the use of a continuous infusion started immediately after administration of the first bolus dose and administered over 4 h.