Ann Surg Oncol DOI 10.1245/s10434-015-4584-2
ORIGINAL ARTICLE – HEPATOBILIARY TUMORS
Perioperative Bundle Decreases Postoperative Hepatic Surgery Infections Maureen V. Hill, MD, Stefan D. Holubar, MD, MS, FACS, FASCRS, Catherine I. Garfield Legare, MHA, Christopher M. Luurtsema, MBA, and Richard J. Barth Jr., MD, FACS Department of Surgery, Dartmouth Hitchcock Medical Center, Lebanon, NH
ABSTRACT Background. Postoperative infections are a costly and morbid complication. The introduction of perioperative infection prevention bundles have decreased surgical site infections (SSIs) in patients undergoing colorectal and pancreatic surgery. Aim. The purpose of this study was to determine if the implementation of a perioperative bundle would reduce postoperative infectious complications in patients undergoing hepatic surgery. Methods. An evidence-based, low cost, perioperative infection bundle was created, and a retrospective review of a prospectively maintained database was performed on 163 consecutive patients undergoing hepatic surgery. Patient characteristics, operative details, outcomes, and complications were reviewed, and differences pre- and post-bundle were assessed with univariate and multivariate analyses. Results. A total of 113 patients received standard infection prophylaxis, while 50 received the perioperative bundle. Twenty-five patients had infections (16 deep abscesses, 3 superficial SSIs, 4 urinary tract infections, 1 pneumonia, 1 bacteremia). The overall infection rate decreased from 20.4 % (23/133) pre-bundle to 4 % (2/50) post-bundle. The SSI rate also decreased from 15 % (17/113) to 4 % (2/50). Univariate analysis showed that institution of the bundle was associated with a lower overall infection rate (p = 0.008), lower SSI rate (p = 0.06), and lower overall complication rate (p = 0.04). Multivariate analysis confirmed that the use of the bundle was independently associated with a lower infection (p = 0.008) and SSI
Ó Society of Surgical Oncology 2015 First Received: 5 February 2015 M. V. Hill, MD e-mail: [email protected]
(p = 0.05) rate. The primary length of stay (LOS) and LOS for 60 days postoperatively both significantly decreased post-bundle (from median of 5–4 days, p B 0.001; 6–4 days, p B 0.001). Conclusions. Implementation of a perioperative infection prevention bundle significantly decreased overall infections, SSIs, and postoperative LOS in patients undergoing hepatic surgery.
Postoperative infections are a morbid and costly complication of any surgical procedure.1–3 Overall, surgical site infections (SSIs) account for approximately 14–16 % of nosocomial infections,4 result in an increased length of stay (LOS),1 and are an independent predictor of postoperative mortality.3 The cost of these infections ranges widely and is dependent on the nature of the infection and the operative procedure. Per infection, the cost can range from a few hundred dollars to tens of thousands of dollars,1,5 and it is estimated that SSIs result in approximately 1 million excess hospital days and more than $1.6 billion in total yearly costs.6 Evidence-based best practices have been implemented in an attempt to decrease the incidence of SSIs.2 Specific perioperative infection prevention bundles have been created focusing on decreasing SSIs in both colorectal and pancreaticoduodenectomy patients.2,7,8 One study showed that with standardized infection prevention care, postoperative SSIs decreased from 14.3 to 8.7 % in patients undergoing colorectal and biliary surgery.2 The colorectal service at the Mayo Clinic, Rochester, was able to reduce their SSI rate from 9.8 to 4 % after initiation of a comprehensive infection prevention bundle.8 Because of these positive impacts on decreasing SSIs, colorectal infection prevention bundles are now being increasingly accepted and incorporated into standard perioperative care. Hepatic resections have been shown to have an SSI rate of 7.7–18.3 %, and overall infection rate of up to
M. V. Hill et al.
21.8 %.3,9,10 Multiple studies have investigated the risk factors associated with SSIs in hepatic resections.10–15 To our knowledge, no studies have been reported showing that specific infection prevention bundles can decrease infectious complications after liver surgery. The purpose of this study was to assess if implementing a perioperative infection prevention bundle would reduce overall infections and SSIs in patients undergoing hepatic surgery. METHODS After a thorough review of the literature, an iterative physician-led consensus building process was used to develop a perioperative hepatic infection prevention bundle (Table 1). Prior to institution of the infection prevention bundle we routinely employed many practices recommended for the control of SSIs.5 Pre-bundle standard practice for liver surgery at our institution included double gloving, hair clipping immediately prior to surgery, povidone-iodine skin preparations, avoidance of drains, use of systemic prophylactic antibiotics within 1 h of surgery, maintenance of intraoperative normothermia, and discontinuation of a urinary catheter on postoperative day 2. If a simultaneous colon resection was performed the patient also received a mechanical and oral antibiotic (flagyl, neomycin) bowel preparation.
a concomitant bowel resection was planned. Patients allergic to penicillin or cefazolin were administered levofloxacin. Another modification of the standard practice was to substitute chlorhexadine-alcohol skin preparation for povidone-iodine.16 Bundle Additions to Standard Practice The following elements were additions to our standard practice. At the time of their preoperative appointment, patients were supplied with chlorhexadine solution and were instructed to wash with it the night prior to and the morning of surgery.4,5 An IobanÒ drape was placed on the skin prior to the start of the procedure.5 We ensured that ampicillin/sulbactam was re-dosed every 3 h during the case.17,18 Prior to case completion and closure, the operative field and wound were washed with a 500 ml solution composed of 1 ml of NeosporinÒ (neomycin sulfate–polymixin B sulfate) GU irrigant and 50,000 units of bacitracin.5,19–21 The total cost of the new/added elements to the bundle per patient was $26.50 (intraoperative ampicillin/sulbactam redosing $3.31, triple antibiotic irrigation $7.76, IobanÒ dressing $9.86, chlorhexadine-alcohol solution and skin preparation $6.25). Bundle Implementation
Bundle Modifications of Standard Practice Our bundle intervention consisted of modification of certain standard elements and the addition of others. Prebundle we used either cefazolin or ampicillin/sulbactam (at approximately equal frequencies) for systemic antibiotic prophylaxis. We chose ampicillin/sulbactam as the standard antibiotic for the bundle since we observed a trend of resistant organisms being isolated from culture data of patients with SSIs who received cefazolin. The combination of cefazolin and metronidazole was used in cases when
The bundle was initiated in December 2012 and applied to 50 consecutive patients undergoing hepatic surgery. Compliance with the bundle was monitored with a checklist that was filled out immediately after each procedure, and confirmed that components of the bundle were followed in all patients. A retrospective review was then performed on these patients as well as 113 previous consecutive historic patients who underwent hepatic surgery from January 2005 through March 2012 by the same surgeon (RJB) prior to implementation of the bundle. Only
TABLE 1 Elements of pre-bundle standard infection reducing care versus bundle care Element
Pre-bundle standard
Preoperative antibiotic
Ampicillin/sulbactam or cefazolin
Ampicillin/sulbactam
Skin preparation
Povidone-iodine
Chlorhexadine
Neosporin–polymixin–bacitracin liver wash and wound wash
N
Y
Chlorhexadine shower night before and morning of surgery
N
Y
Ioban drape
N
Y
Redosing of antibiotics every 3 h
N
Y
Avoidance of drains
Y
Y
Urinary catheter discontinuation by POD 2
Y
Y
If concomitant colorectal resection, mechanical and oral antibiotic preparation
Y
Y
Ò
POD postoperative day, Y yes, N no
Bundle
Perioperative Bundle Decreases Postoperative Hepatic Surgery Infections
patients who underwent primary resection by this surgeon were included in the review. A prospectively reviewed liver surgery database was maintained and included patient demographics, operative details and postoperative outcomes. Primary LOS for the initial hospitalization and total LOS (TLOS), which included readmitted days within the 60 days after surgical resection, were recorded. If the patient developed an infectious complication, culture data was reviewed, as were the bacteria resistive patterns. Over the period of time included in this study, we evaluated the effect of a low-calorie, low-fat diet on hepatic steatosis, reported elsewhere.22 Some of the patients followed this diet for 1 week prior to surgery. Around the same time we initiated the infection prevention bundle, we began a prospective, randomized study of the effect of a low-calorie, low-fat diet for 1 week prior to surgery in patients with a body mass index (BMI) [25. Statistical Considerations Our primary endpoint was the overall infection rate, and our secondary endpoint was the combined (superficial, deep, organ space) SSI rate. Other endpoints included overall complications, readmission rate, LOS and TLOS in days. Univariate analysis of continuous and categorical data, respectively, was assessed using parametric (Students’ t test, Chi square analysis) and non-parametric (Fisher’s exact test, Wilcoxon rank-sum test), and analyzed for potential predictors of postoperative overall infections
or SSIs. A multivariate logistic regression model was then used to assess independent predictors of infection and SSI. Bundle was the primary predictor and was adjusted for confounding by variables that differed between the preand post-bundle groups at baseline (Table 2), as well as other potential independent predictors of infection. All statistical analyses were performed using JMP version 10 (SAS Institute, Cary, NC, USA) for Mac. RESULTS Patient characteristics of the pre- and post-bundle patient groups were similar (Table 2). Preoperative ampicillin/sulbactam use was lower in the pre-bundle group (42 vs. 80 %, p \ 0.001). Patients in the bundle group who did not receive ampicillin/sulbactam either had a concomitant colon resection or were allergic to ampicillin. Preoperative diet use was higher in the pre-bundle group (47 vs. 21 %, p = 0.003), and intraoperative blood loss was slightly higher in the pre-bundle group (754 vs. 593, p = 0.04). The overall postoperative infection rate decreased from 20.4 % (23/113) before the bundle to 4 % (2/50) after the bundle was initiated (p = 0.008) (Table 3). Pre-bundle infectious complications included 17 SSIs (14 deep hepatic abscesses, 3 superficial SSIs), 4 urinary tract infections, 1 pneumonia, and 1 bacteremia of unknown origin. Both infections in the post-bundle group were hepatic abscesses. In univariate analysis of perioperative variables potentially
TABLE 2 Patient characteristics pre and post infection prevention bundle Pre-bundle (n = 113)
Post-bundle (n = 50)
Univariate p value
Age, years (mean ± SD)
60.2 ± 11.8
58.4 ± 11.6
[0.2
Female sex
49 (43.4)
23 (46)
[0.2
Body mass index, kg/m2 (mean ± SD) Diabetic
27.4 ± 6.1 14 (12.4)
26.0 ± 4.7 8 (16)
0.16 [0.2
Malignant diagnosis
107 (94.7)
48 (96)
[0.2
Preoperative chemotherapy
68 (60.2)
29 (58)
[0.2
Preoperative diet
53 (46.9)
11 (22)
Preoperative antibiotics
113 (100)
50 (100)
Preoperative ampicillin/sulbactam
48 (42.5)
40 (80)
Lobectomy
21 (18.8)
10 (20)
1–2 Segments
64 (57.1)
32 (64)
– \0.001a [0.2
Extent of resection
C3 Segments
0.003b
27 (24.1)
8 (16)
Bowel resection
17 (15)
11 (22)
Blood loss (ml) (mean ± SD)
754 ± 531
593 ± 633
Transfusion
8 (7.1)
2 (4)
Values represent mean ± SD or frequency (percent) a
Not included in model due to colinearity with bundle
b
Included in multivariate model
[0.2 0.04b [0.2
M. V. Hill et al.
associated with overall postoperative infections (Table 4), lack of the bundle, male sex, preoperative diet, and diabetes were potential predictors of overall infections. Using multivariate analysis, adjusting for sex, diabetes, preoperative diet, and blood loss, we found that institution of the bundle was the only independent predictor of reduced overall infections [p = 0.008, odds ratio (OR) 0.17, 95 % CI 0.03–0.66]. SSIs decreased from 17/113 (15 %) to 2/50 (4 %) (p = 0.06). In univariate analysis of perioperative variables potentially associated with SSIs (Table 4), lack of the bundle, male sex, and diabetes were potential predictors. In multivariate analysis, adjusting for sex, diabetes, blood loss, and preoperative diet, we found that lack of the bundle (p = 0.05, OR 4.11, 95 % CI 1.007–28.4), male sex (p = 0.02, OR 4.25, 95 % CI 1.28–19.4), and diabetes (p = 0.03, OR 4.33, 95 % CI 1.11–16.3) were significant predictors of SSIs. The overall complication rate decreased from 33.7 % pre-bundle to 18 % post-bundle (p = 0.04) (Table 3). There was no significant difference in the rate of non-infectious complications in the pre-bundle (19.5 %, 22/113) compared with the post-bundle patients (10 %, 5/50). Noninfectious complications included atrial fibrillation/arrhythmias, diarrhea, ileus, small bowel obstruction, sterile intra-abdominal fluid collections, sterile bile leak, failure to thrive, myocardial infarction, non-ST elevation myocardial infarction (NSTEMI), and urinary retention. There was one mortality in the pre-bundle group and none in the postbundle group during the 60-day follow-up period. Postoperative median LOS decreased from 5 [interquartile range (IQR) 4.5–7] to 4 (IQR 3–5) days (p \ 0.001) after the bundle was initiated (Table 3). Median TLOS in the 60-day postoperative period decreased from 6 (IQR 5–7) to 4 (IQR 3–5) days (p B 0.001). The readmission rate decreased from 13.3 to 6 % (Table 3). The inpatient costs of all patients for the 60-days after surgery was calculated from administrative databases. The overall mean cost per inpatient day was $1543.
An antibiogram of the bacteria isolated from culture data of patients with deep SSIs was collected (Table 5). Eight patients had infections with multiple organisms. Gram-positive, gram-negative, and anaerobes were isolated in 41, 52, and 7 % of the cultures, respectively. Data on the prophylactic antibiotic was also collected. A high proportion of patients who received either cephalexin (67 %) or ampicillin/sulbactam (50 %) grew resistant bacteria in their SSIs. DISCUSSION Our overall infection and SSI rates for patients undergoing hepatic surgery prior to initiation of the infection prevention bundle were consistent with the literature.3,9,10 We have now demonstrated that implementation of an infection prevention bundle in hepatic surgery patients can significantly decrease overall postoperative infections and SSIs. By reducing infectious complications (mainly deep abscesses), we decreased the postoperative LOS by 1 day and the TLOS by 2 days. Since the added cost of institution of the bundle was minimal ($26), its use resulted in substantial cost savings. At an average inpatient cost of $1543 per day, a decrease in TLOS of 2 days per patient resulted in an average savings of $3086 per patient with implementation of the bundle. As shown in Table 2, there were few significant differences in the clinical characteristics of the pre- and postbundle patients; specifically, a lower percentage of patients in the post-bundle group were on the preoperative lowcalorie, low-fat diet and a higher proportion received preoperative ampicillin/sulbactam. Since the preoperative diet was not associated with a significantly higher infection rate (Table 4), this difference in baseline patient characteristics did not confound the observed relationship between the bundle and observed infection rates. Furthermore, diet status was adjusted for in the multivariate model. Since ampicillin/sulbactam was the standard antibiotic used in our post-bundle population, it was expected that a higher
TABLE 3 Short-term (60-day) postoperative outcomes Pre-bundle (n = 113)
Post-bundle (n = 50)
p value
Infectious complications
23 (20.4)
2 (4)
0.008
SSIs
17 (15)
2 (4)
0.06
Overall complications Primary LOS, days [median (IQR)]
37 (33.7) 5 (4.5–7)
9 (18) 4 (3–5)
0.04 \0.001 \0.001
Total LOS, days [median (IQR)]
6 (5–7)
4 (3–5)
Readmissions
15 (13.3)
3 (6)
Mortality
1 (1)
Data are expressed as n (%) unless otherwise specified NS not significant, LOS length of stay, IQR interquartile range, SSIs surgical site infections
0
NS NS
Perioperative Bundle Decreases Postoperative Hepatic Surgery Infections TABLE 4 Univariate analysis of parameters associated with postoperative hepatic surgery infections Surgical site infections (19 patients)
Overall infections (25 patients)
No. infected/total [rate] (%)
Univariate p value
Multivariate p value
No. infected/total [rate] (%)
Univariate p value
Multivariate p value
Pre-bundle
17/113 (15)
0.06a
0.05
23/113 (20.4)
0.008a
0.008
Post-bundle
2/50 (4)
Preoperative diet No preoperative diet Female Male Diabetic Non-diabetic
10/64 (15.6) 9/99 (9.1)
0.22
NS
13/64 (20.3) 12/99 (12.1)
0.18a
NS
3/72 (4.2)
0.01a
0.02
6/72 (8.3)
0.03a
NS
0.03a
0.03
0.11a
NS
16/91 (17.6) 6/22 (27.3) 10/84 (11.9)
Age \ 60 years
9/79 (11.4)
BMI C 26
9/82 (10.9)
BMI \ 26
No preoperative chemotherapy Pringle maneuver No pringle maneuver Bowel resection
13/84 (15.5)
7/67 (10.5) 12/96 (12.5) 4/28 (14.3) 15/135 (11.1)
Blood loss C600 cc
10/79 (12.6)
Blood loss \600 cc
8/80 (10)
Transfusion
1/10 (10)
[0.2
12/79 (15.2) [0.2
11/82 (13.4)
[0.2
25/155 (16.1)
[0.2
14/81 (17.3) [0.2
0/8 (0) [0.2
15/97 (15.5)
9/66 (13.6)
No bowel resection
No transfusion
[0.2
0/8 (0) 10/97 (10.3)
6/22 (27.3) 19/141 (13.5)
10/81 (12.4) 19/155 (12.3)
Benign diagnosis Preoperative chemotherapy
19/91 (20.9)
13/141 (9.2)
Age C 60 years
Malignant diagnosis
2/50 (4)
[0.2
10/66 (15.2) [0.2
9/67 (13.4)
[0.2
[0.2
16/96 (16.7) 5/28 (17.9)
[0.2
20/135 (14.8) [0.2a
NS
14/79 (17.7)
[0.2a
NS
10/80 (12.5) [0.2
18/153 (11.8)
2/10 (20)
[0.2
23/153 (15)
BMI body mass index, NS not significant a
Included in the multivariate model
proportion of bundle patients received this antibiotic compared with the pre-bundle group. The relative contribution of the individual elements of the bundle to the improved outcome is unclear. Some authors recommend cefazolin as the preoperative antibiotic for hepatic surgery.5,23 The choice of ampicillin/sulbactam as the preoperative antibiotic in our bundle was based on our observations, prior to bundle initiation, that many organisms growing from postoperative abscess cultures were resistant to cefazolin. However, at the conclusion of the study we noted that patients commonly grew organisms from abscesses that were resistant to their respective preoperative antibiotic (Table 5). Therefore, we feel it is unlikely that the use of ampicillin/sulbactam was a major factor in the decreased postoperative infection rate. It is more likely that the other elements of the bundle contributed, to a greater degree, to the decrease in infection rate. Chlorhexadine showers were added to the bundle because it has been established that skin bacteria counts are decreased with chlorhexadine application.5 Although a
meta-analysis questions the effectiveness of preoperative bathing with antiseptic solutions,24 a clinical trial demonstrated decreased SSIs with chlorhexadine bathing.25 We chose chlorhexadine, rather than povidone-iodine, for the skin preparation because randomized, prospective studies showed decreased local positive culture results and decreased SSIs after surgery.5,15 We included IobanÒ drapes in the bundle because a clinical trial evaluating the use of an iodophor impregnated skin drape in hepatic surgery demonstrated a lower SSI rate with use.26 Standardized antibiotic redosing was implemented based on the specific drug half-life because it has been shown that therapeutic drug levels at the time of wound closure are critical determinants of SSI rates.18 There is extensive data on the use of a topical antimicrobial in preventing SSIs. Clinical trials have shown a decrease in SSI rates in both biliary surgery as well as surgery localized to the subcutaneous tissue 19–21 The decision was therefore made to add the broad-spectrum neomycin–polymixin–bacitracin wound irrigation to our bundle.
M. V. Hill et al. TABLE 5 Antibiogram: culture data from deep SSI and sensitivities to preoperative antibiotics Prophylactic cephalexin
Prophylactic ampicillin/ sulbactam
Gram-positive organisms Corynebacterium
NA
Coagulase-negative staphylococci 1
R
2
R
MRSA
R
Propionibacterium acnes 1
NA
2 Streptococci viridans
S S
Enterococcus 1
R
2 3 4
S S R
Gram-negative organisms Serratia
R
Escherichia coli 1
S
2
S
CONCLUSIONS
3
S
4
S
Pseudomonas 1
S
2 Citrobacter
R R
Enterobacter 1
R R
2
R
3
R
Morganella morganii
R
Anaerobes Bacteroides fragilis
S
Clostridium perfringens Any resistant organisms
We were able to show that a low-cost, easy to implement bundle was able to decrease not only postoperative infection rates but also total complications, LOS, and cost. This suggests that such bundles should be routinely developed and implemented in order to optimize outcomes in this at-risk population.
R
2 Klebsiella 1
preparation.7 The colorectal bundle of the Mayo Clinic, Rochester, included chlorhexadine preoperative washes, chlorhexadine preparation, and routine antibiotic redosing.8 This intervention study, like other studies of infection prevention bundles,7,8 lacks the rigor of a randomized, prospective trial in determining the value of an infection prevention bundle. However, the strengths of the study include the uniformity of the patients treated and the surgical techniques used pre- and post-bundle; all patients were treated by a single surgeon, at a single institution. Uniform compliance with the bundle interventions was another study strength.27 Other investigators have reported that diabetes, obesity, and excessive blood loss are risk factors for postoperative infections in liver surgery patients.10,12,13 In our multivariate analysis, we also found that diabetes was an independent predictor of post-operative infections. Although our diabetic patients underwent routine blood glucose checks and were treated with sliding-scale insulin, we did not institute a strict regimen for glucose control. Such regimens have recently been shown to be effective in reducing SSIs in diabetic patients undergoing surgery, and will be considered as an addition to our bundle in the future.28
NA 8/12 (66.7 %)
6/12 (50 %)
Numbers represent separate isolates of bacteria R resistant, S sensitive, NA no sensitivities reported, MRSA methicillin-resistant Staphylococcus aureus
The content of our bundle shared some elements utilized in other infection prevention bundles. The pancreaticoduodenectomy bundle created by Lavu et al. contained preoperative chlorhexadine wipes and chlorhexadine
REFERENCES 1. Coello R, Charlett A, Wilson J, Ward V, Pearson A, Boriello P. Adverse impact of surgical site infections in English hospital. J Hosp Infect. 2005;60:93–103. 2. Forbes S, Stephen W, Harper W, Loeb M, Smith R, Christofferson E, et al. Implementation of evidence-based practices for surgical site infection prophylaxis: results of a pre- and postintervention study. J Am Coll Surg. 2008;207:336–341. 3. Astagneau P, Rious C, Golliot F, Brucker G. Morbidity and mortality associated with surgical site infections: results from the 1997-1999 INCISO surveillance. J Hosp Infect. 2001;48:267–274. 4. Mangram AJ, Horan TC, Pearson ML, Silver LC, Jarvis WR. Guideline for prevention of surgical site infection, 1999. Am J Infect Control. 1999;27:97–134. 5. Alexander JW, Solomkin J, Edwards M. Updated recommendations for control of surgical site infections. Ann Surg. 2011;253:1082–1093. 6. de Lissovoy G, Fraeman K, Hutchins V, Murphy D, Song D, Vaughn B. Surgical site infection: incidence and impact of hospital utilization and treatment costs. Am J Infect Control. 2009;37:387–397.
Perioperative Bundle Decreases Postoperative Hepatic Surgery Infections 7. Lavu H, Klinge M, Nowcid L, Cohn H, Grenda D, Sauter P, et al. Perioperative surgical care bundle reduces pancreaticoduodenectomy wound infections. J Surg Res. 2012;174:215–221. 8. Cima R, Dankbar E, Lovely J, Pendlimari R, Aronhalt K, Nehring S, et al. Colorectal surgery surgical site infection reduction program: a national surgical quality improvement program-driven multidisciplinary single-institution experience. J Am Coll Surg. 2013;216:23–33. 9. Elola-Olaso AM, Davenport DL, Hundley JC, Daily MF, Gedaly R. Predictors of surgical site infection after liver resection: a multicentre analysis using national surgical quality improvement program data. HPB (Oxford). 2012;14:136–141. 10. Okabayashi T, Nishimori I, Yamashita K, Sugimoto T, Yatabe T, Maeda H, et al. Risk factors and predictors for surgical site infection after hepatic resection. J Hosp Infect. 2009;73:47–53. 11. Ceppa E, Pitt H, House M, Kilbane M, Nakeeb A, Schmidt C, et al. Reducing surgical site infections in hepatopancreatobiliary surgery. HPB (Oxford). 2013;15:384–391. 12. Gedaly R, McHugh P, Johnston T, Jeon H, Ranjan D, Davenport D. Obesity, diabetes, and smoking are important determinants of resource utilization in liver resection: a multicenter analysis of 1029 patients. Ann Surg. 2009;249(3):414–419. 13. Kobayashi S, Gotohda N, Nakagohri T, Takahashi S, Konishi M, Kinoshita T. Risk factors of surgical site infection after hepatectomy for liver cancers. World J Surg. 2009;33:312–317. 14. Moreno Elola-Olaso A, Davenport D, Hundley J, Daily M, Gedaly R. Predictors of surgical site infection after liver resection: a multicentre analysis using national surgical quality improvement program data. HPB (Oxford). 2012;14:135–141. 15. Pessaux P, van den Broek M, Wu T, Olde Damink S, Dejong C, Ntourakis D, et al. Identification and validation of risk factors for postoperative infectious complications following hepatectomy. J Gastrointest Surg. 2013;17:1907–1916. 16. Bibbo C, Patel D, Gehrmann R, Lin S. Chlorhexadine provides superior skin decontamination in foot and ankle surgery. Clin Orthop Relat Res. 2005;438:204–208. 17. Hawn M, Itani K, Gray S, Vick C, Henderson W, Houston T. Association of timely administration of prophylactic antibiotics for major surgical procedures and surgical site infection. J Am Coll Surg. 2008;206:815–821.
18. Zelenitsky S, Ariano R, Harding G, Silverman R. Antibiotic pharmacodynamics in surgical prophylaxis: an association between intraoperative antibiotic concentrations and efficacy. Antimicrob Agents Chemother. 2002;46(9):3026–3030. 19. Lord JW, LaRaja RD, Daliana M, Gordon MT. Prophylactic antibiotic wound irrigation in gastric, biliary and colonic surgery. Am J Surg. 1983;145(2):209–212. 20. Lord J, Rossi G, Daliana M. Intraoperative antibiotic wound lavage. Ann Surg. 1977;185(6):634–641. 21. Halasz NA. Wound infection and topical antibiotics. Arch Surg. 1977;112:1240–1244. 22. Reeves JG, Suriawinata AA, Ng DP, Holubar SD, Mills JB, Barth RJ. Short-term preoperative diet modification reduces steatosis and blood loss in patients undergoing liver resection. Surgery. 2013;154(11):1031–1037. 23. Bratzler D, Houck P. Antimicrobial prophylaxis for surgery: an advisory statement from the national surgical infection prevention project. Am J Surg. 2005;189:395–404. 24. Webster J, Osborne S. Preoperative bathing or showering with skin antiseptics to prevent surgical site infection. Cochrane Database Syst Rev. 2012;9:CD004985. 25. Wihlborg O. The effect of washing with chlorhexadine soap on wound infection rate in general surgery. A controlled clinical study. Ann Chir Gynaecol. 1987;76(5):263–5. 26. Yoshimura Y, Kubo S, Hirohashi K, Ogawa M. Plastic iodophor drape during liver surgery operative use of the iodophor-impregnated adhesive drape to prevent wound infection during high risk surgery. World J Surg. 2003;27:685–8. 27. Waits SA, Fritze D, Banerjee M, Zhang W, Kubus J, Englesbe MJ, et al. Developing an argument for bundled interventions to reduce surgical site infection in colorectal surgery. Surgery. 2014;115:602–606. 28. Wallaert JB, Nolan BW, Adams J, Stanley A, Eldrup-Jorgensen J, Cronenwett JL, et al. The impact of diabetes on postoperative outcomes following lower-extremity bypass surgery. J Vasc Surg. 2012;56:1317–1323.
ORIGINAL ARTICLE – HEPATOBILIARY TUMORS
Perioperative Bundle Decreases Postoperative Hepatic Surgery Infections Maureen V. Hill, MD, Stefan D. Holubar, MD, MS, FACS, FASCRS, Catherine I. Garfield Legare, MHA, Christopher M. Luurtsema, MBA, and Richard J. Barth Jr., MD, FACS Department of Surgery, Dartmouth Hitchcock Medical Center, Lebanon, NH
ABSTRACT Background. Postoperative infections are a costly and morbid complication. The introduction of perioperative infection prevention bundles have decreased surgical site infections (SSIs) in patients undergoing colorectal and pancreatic surgery. Aim. The purpose of this study was to determine if the implementation of a perioperative bundle would reduce postoperative infectious complications in patients undergoing hepatic surgery. Methods. An evidence-based, low cost, perioperative infection bundle was created, and a retrospective review of a prospectively maintained database was performed on 163 consecutive patients undergoing hepatic surgery. Patient characteristics, operative details, outcomes, and complications were reviewed, and differences pre- and post-bundle were assessed with univariate and multivariate analyses. Results. A total of 113 patients received standard infection prophylaxis, while 50 received the perioperative bundle. Twenty-five patients had infections (16 deep abscesses, 3 superficial SSIs, 4 urinary tract infections, 1 pneumonia, 1 bacteremia). The overall infection rate decreased from 20.4 % (23/133) pre-bundle to 4 % (2/50) post-bundle. The SSI rate also decreased from 15 % (17/113) to 4 % (2/50). Univariate analysis showed that institution of the bundle was associated with a lower overall infection rate (p = 0.008), lower SSI rate (p = 0.06), and lower overall complication rate (p = 0.04). Multivariate analysis confirmed that the use of the bundle was independently associated with a lower infection (p = 0.008) and SSI
Ó Society of Surgical Oncology 2015 First Received: 5 February 2015 M. V. Hill, MD e-mail: [email protected]
(p = 0.05) rate. The primary length of stay (LOS) and LOS for 60 days postoperatively both significantly decreased post-bundle (from median of 5–4 days, p B 0.001; 6–4 days, p B 0.001). Conclusions. Implementation of a perioperative infection prevention bundle significantly decreased overall infections, SSIs, and postoperative LOS in patients undergoing hepatic surgery.
Postoperative infections are a morbid and costly complication of any surgical procedure.1–3 Overall, surgical site infections (SSIs) account for approximately 14–16 % of nosocomial infections,4 result in an increased length of stay (LOS),1 and are an independent predictor of postoperative mortality.3 The cost of these infections ranges widely and is dependent on the nature of the infection and the operative procedure. Per infection, the cost can range from a few hundred dollars to tens of thousands of dollars,1,5 and it is estimated that SSIs result in approximately 1 million excess hospital days and more than $1.6 billion in total yearly costs.6 Evidence-based best practices have been implemented in an attempt to decrease the incidence of SSIs.2 Specific perioperative infection prevention bundles have been created focusing on decreasing SSIs in both colorectal and pancreaticoduodenectomy patients.2,7,8 One study showed that with standardized infection prevention care, postoperative SSIs decreased from 14.3 to 8.7 % in patients undergoing colorectal and biliary surgery.2 The colorectal service at the Mayo Clinic, Rochester, was able to reduce their SSI rate from 9.8 to 4 % after initiation of a comprehensive infection prevention bundle.8 Because of these positive impacts on decreasing SSIs, colorectal infection prevention bundles are now being increasingly accepted and incorporated into standard perioperative care. Hepatic resections have been shown to have an SSI rate of 7.7–18.3 %, and overall infection rate of up to
M. V. Hill et al.
21.8 %.3,9,10 Multiple studies have investigated the risk factors associated with SSIs in hepatic resections.10–15 To our knowledge, no studies have been reported showing that specific infection prevention bundles can decrease infectious complications after liver surgery. The purpose of this study was to assess if implementing a perioperative infection prevention bundle would reduce overall infections and SSIs in patients undergoing hepatic surgery. METHODS After a thorough review of the literature, an iterative physician-led consensus building process was used to develop a perioperative hepatic infection prevention bundle (Table 1). Prior to institution of the infection prevention bundle we routinely employed many practices recommended for the control of SSIs.5 Pre-bundle standard practice for liver surgery at our institution included double gloving, hair clipping immediately prior to surgery, povidone-iodine skin preparations, avoidance of drains, use of systemic prophylactic antibiotics within 1 h of surgery, maintenance of intraoperative normothermia, and discontinuation of a urinary catheter on postoperative day 2. If a simultaneous colon resection was performed the patient also received a mechanical and oral antibiotic (flagyl, neomycin) bowel preparation.
a concomitant bowel resection was planned. Patients allergic to penicillin or cefazolin were administered levofloxacin. Another modification of the standard practice was to substitute chlorhexadine-alcohol skin preparation for povidone-iodine.16 Bundle Additions to Standard Practice The following elements were additions to our standard practice. At the time of their preoperative appointment, patients were supplied with chlorhexadine solution and were instructed to wash with it the night prior to and the morning of surgery.4,5 An IobanÒ drape was placed on the skin prior to the start of the procedure.5 We ensured that ampicillin/sulbactam was re-dosed every 3 h during the case.17,18 Prior to case completion and closure, the operative field and wound were washed with a 500 ml solution composed of 1 ml of NeosporinÒ (neomycin sulfate–polymixin B sulfate) GU irrigant and 50,000 units of bacitracin.5,19–21 The total cost of the new/added elements to the bundle per patient was $26.50 (intraoperative ampicillin/sulbactam redosing $3.31, triple antibiotic irrigation $7.76, IobanÒ dressing $9.86, chlorhexadine-alcohol solution and skin preparation $6.25). Bundle Implementation
Bundle Modifications of Standard Practice Our bundle intervention consisted of modification of certain standard elements and the addition of others. Prebundle we used either cefazolin or ampicillin/sulbactam (at approximately equal frequencies) for systemic antibiotic prophylaxis. We chose ampicillin/sulbactam as the standard antibiotic for the bundle since we observed a trend of resistant organisms being isolated from culture data of patients with SSIs who received cefazolin. The combination of cefazolin and metronidazole was used in cases when
The bundle was initiated in December 2012 and applied to 50 consecutive patients undergoing hepatic surgery. Compliance with the bundle was monitored with a checklist that was filled out immediately after each procedure, and confirmed that components of the bundle were followed in all patients. A retrospective review was then performed on these patients as well as 113 previous consecutive historic patients who underwent hepatic surgery from January 2005 through March 2012 by the same surgeon (RJB) prior to implementation of the bundle. Only
TABLE 1 Elements of pre-bundle standard infection reducing care versus bundle care Element
Pre-bundle standard
Preoperative antibiotic
Ampicillin/sulbactam or cefazolin
Ampicillin/sulbactam
Skin preparation
Povidone-iodine
Chlorhexadine
Neosporin–polymixin–bacitracin liver wash and wound wash
N
Y
Chlorhexadine shower night before and morning of surgery
N
Y
Ioban drape
N
Y
Redosing of antibiotics every 3 h
N
Y
Avoidance of drains
Y
Y
Urinary catheter discontinuation by POD 2
Y
Y
If concomitant colorectal resection, mechanical and oral antibiotic preparation
Y
Y
Ò
POD postoperative day, Y yes, N no
Bundle
Perioperative Bundle Decreases Postoperative Hepatic Surgery Infections
patients who underwent primary resection by this surgeon were included in the review. A prospectively reviewed liver surgery database was maintained and included patient demographics, operative details and postoperative outcomes. Primary LOS for the initial hospitalization and total LOS (TLOS), which included readmitted days within the 60 days after surgical resection, were recorded. If the patient developed an infectious complication, culture data was reviewed, as were the bacteria resistive patterns. Over the period of time included in this study, we evaluated the effect of a low-calorie, low-fat diet on hepatic steatosis, reported elsewhere.22 Some of the patients followed this diet for 1 week prior to surgery. Around the same time we initiated the infection prevention bundle, we began a prospective, randomized study of the effect of a low-calorie, low-fat diet for 1 week prior to surgery in patients with a body mass index (BMI) [25. Statistical Considerations Our primary endpoint was the overall infection rate, and our secondary endpoint was the combined (superficial, deep, organ space) SSI rate. Other endpoints included overall complications, readmission rate, LOS and TLOS in days. Univariate analysis of continuous and categorical data, respectively, was assessed using parametric (Students’ t test, Chi square analysis) and non-parametric (Fisher’s exact test, Wilcoxon rank-sum test), and analyzed for potential predictors of postoperative overall infections
or SSIs. A multivariate logistic regression model was then used to assess independent predictors of infection and SSI. Bundle was the primary predictor and was adjusted for confounding by variables that differed between the preand post-bundle groups at baseline (Table 2), as well as other potential independent predictors of infection. All statistical analyses were performed using JMP version 10 (SAS Institute, Cary, NC, USA) for Mac. RESULTS Patient characteristics of the pre- and post-bundle patient groups were similar (Table 2). Preoperative ampicillin/sulbactam use was lower in the pre-bundle group (42 vs. 80 %, p \ 0.001). Patients in the bundle group who did not receive ampicillin/sulbactam either had a concomitant colon resection or were allergic to ampicillin. Preoperative diet use was higher in the pre-bundle group (47 vs. 21 %, p = 0.003), and intraoperative blood loss was slightly higher in the pre-bundle group (754 vs. 593, p = 0.04). The overall postoperative infection rate decreased from 20.4 % (23/113) before the bundle to 4 % (2/50) after the bundle was initiated (p = 0.008) (Table 3). Pre-bundle infectious complications included 17 SSIs (14 deep hepatic abscesses, 3 superficial SSIs), 4 urinary tract infections, 1 pneumonia, and 1 bacteremia of unknown origin. Both infections in the post-bundle group were hepatic abscesses. In univariate analysis of perioperative variables potentially
TABLE 2 Patient characteristics pre and post infection prevention bundle Pre-bundle (n = 113)
Post-bundle (n = 50)
Univariate p value
Age, years (mean ± SD)
60.2 ± 11.8
58.4 ± 11.6
[0.2
Female sex
49 (43.4)
23 (46)
[0.2
Body mass index, kg/m2 (mean ± SD) Diabetic
27.4 ± 6.1 14 (12.4)
26.0 ± 4.7 8 (16)
0.16 [0.2
Malignant diagnosis
107 (94.7)
48 (96)
[0.2
Preoperative chemotherapy
68 (60.2)
29 (58)
[0.2
Preoperative diet
53 (46.9)
11 (22)
Preoperative antibiotics
113 (100)
50 (100)
Preoperative ampicillin/sulbactam
48 (42.5)
40 (80)
Lobectomy
21 (18.8)
10 (20)
1–2 Segments
64 (57.1)
32 (64)
– \0.001a [0.2
Extent of resection
C3 Segments
0.003b
27 (24.1)
8 (16)
Bowel resection
17 (15)
11 (22)
Blood loss (ml) (mean ± SD)
754 ± 531
593 ± 633
Transfusion
8 (7.1)
2 (4)
Values represent mean ± SD or frequency (percent) a
Not included in model due to colinearity with bundle
b
Included in multivariate model
[0.2 0.04b [0.2
M. V. Hill et al.
associated with overall postoperative infections (Table 4), lack of the bundle, male sex, preoperative diet, and diabetes were potential predictors of overall infections. Using multivariate analysis, adjusting for sex, diabetes, preoperative diet, and blood loss, we found that institution of the bundle was the only independent predictor of reduced overall infections [p = 0.008, odds ratio (OR) 0.17, 95 % CI 0.03–0.66]. SSIs decreased from 17/113 (15 %) to 2/50 (4 %) (p = 0.06). In univariate analysis of perioperative variables potentially associated with SSIs (Table 4), lack of the bundle, male sex, and diabetes were potential predictors. In multivariate analysis, adjusting for sex, diabetes, blood loss, and preoperative diet, we found that lack of the bundle (p = 0.05, OR 4.11, 95 % CI 1.007–28.4), male sex (p = 0.02, OR 4.25, 95 % CI 1.28–19.4), and diabetes (p = 0.03, OR 4.33, 95 % CI 1.11–16.3) were significant predictors of SSIs. The overall complication rate decreased from 33.7 % pre-bundle to 18 % post-bundle (p = 0.04) (Table 3). There was no significant difference in the rate of non-infectious complications in the pre-bundle (19.5 %, 22/113) compared with the post-bundle patients (10 %, 5/50). Noninfectious complications included atrial fibrillation/arrhythmias, diarrhea, ileus, small bowel obstruction, sterile intra-abdominal fluid collections, sterile bile leak, failure to thrive, myocardial infarction, non-ST elevation myocardial infarction (NSTEMI), and urinary retention. There was one mortality in the pre-bundle group and none in the postbundle group during the 60-day follow-up period. Postoperative median LOS decreased from 5 [interquartile range (IQR) 4.5–7] to 4 (IQR 3–5) days (p \ 0.001) after the bundle was initiated (Table 3). Median TLOS in the 60-day postoperative period decreased from 6 (IQR 5–7) to 4 (IQR 3–5) days (p B 0.001). The readmission rate decreased from 13.3 to 6 % (Table 3). The inpatient costs of all patients for the 60-days after surgery was calculated from administrative databases. The overall mean cost per inpatient day was $1543.
An antibiogram of the bacteria isolated from culture data of patients with deep SSIs was collected (Table 5). Eight patients had infections with multiple organisms. Gram-positive, gram-negative, and anaerobes were isolated in 41, 52, and 7 % of the cultures, respectively. Data on the prophylactic antibiotic was also collected. A high proportion of patients who received either cephalexin (67 %) or ampicillin/sulbactam (50 %) grew resistant bacteria in their SSIs. DISCUSSION Our overall infection and SSI rates for patients undergoing hepatic surgery prior to initiation of the infection prevention bundle were consistent with the literature.3,9,10 We have now demonstrated that implementation of an infection prevention bundle in hepatic surgery patients can significantly decrease overall postoperative infections and SSIs. By reducing infectious complications (mainly deep abscesses), we decreased the postoperative LOS by 1 day and the TLOS by 2 days. Since the added cost of institution of the bundle was minimal ($26), its use resulted in substantial cost savings. At an average inpatient cost of $1543 per day, a decrease in TLOS of 2 days per patient resulted in an average savings of $3086 per patient with implementation of the bundle. As shown in Table 2, there were few significant differences in the clinical characteristics of the pre- and postbundle patients; specifically, a lower percentage of patients in the post-bundle group were on the preoperative lowcalorie, low-fat diet and a higher proportion received preoperative ampicillin/sulbactam. Since the preoperative diet was not associated with a significantly higher infection rate (Table 4), this difference in baseline patient characteristics did not confound the observed relationship between the bundle and observed infection rates. Furthermore, diet status was adjusted for in the multivariate model. Since ampicillin/sulbactam was the standard antibiotic used in our post-bundle population, it was expected that a higher
TABLE 3 Short-term (60-day) postoperative outcomes Pre-bundle (n = 113)
Post-bundle (n = 50)
p value
Infectious complications
23 (20.4)
2 (4)
0.008
SSIs
17 (15)
2 (4)
0.06
Overall complications Primary LOS, days [median (IQR)]
37 (33.7) 5 (4.5–7)
9 (18) 4 (3–5)
0.04 \0.001 \0.001
Total LOS, days [median (IQR)]
6 (5–7)
4 (3–5)
Readmissions
15 (13.3)
3 (6)
Mortality
1 (1)
Data are expressed as n (%) unless otherwise specified NS not significant, LOS length of stay, IQR interquartile range, SSIs surgical site infections
0
NS NS
Perioperative Bundle Decreases Postoperative Hepatic Surgery Infections TABLE 4 Univariate analysis of parameters associated with postoperative hepatic surgery infections Surgical site infections (19 patients)
Overall infections (25 patients)
No. infected/total [rate] (%)
Univariate p value
Multivariate p value
No. infected/total [rate] (%)
Univariate p value
Multivariate p value
Pre-bundle
17/113 (15)
0.06a
0.05
23/113 (20.4)
0.008a
0.008
Post-bundle
2/50 (4)
Preoperative diet No preoperative diet Female Male Diabetic Non-diabetic
10/64 (15.6) 9/99 (9.1)
0.22
NS
13/64 (20.3) 12/99 (12.1)
0.18a
NS
3/72 (4.2)
0.01a
0.02
6/72 (8.3)
0.03a
NS
0.03a
0.03
0.11a
NS
16/91 (17.6) 6/22 (27.3) 10/84 (11.9)
Age \ 60 years
9/79 (11.4)
BMI C 26
9/82 (10.9)
BMI \ 26
No preoperative chemotherapy Pringle maneuver No pringle maneuver Bowel resection
13/84 (15.5)
7/67 (10.5) 12/96 (12.5) 4/28 (14.3) 15/135 (11.1)
Blood loss C600 cc
10/79 (12.6)
Blood loss \600 cc
8/80 (10)
Transfusion
1/10 (10)
[0.2
12/79 (15.2) [0.2
11/82 (13.4)
[0.2
25/155 (16.1)
[0.2
14/81 (17.3) [0.2
0/8 (0) [0.2
15/97 (15.5)
9/66 (13.6)
No bowel resection
No transfusion
[0.2
0/8 (0) 10/97 (10.3)
6/22 (27.3) 19/141 (13.5)
10/81 (12.4) 19/155 (12.3)
Benign diagnosis Preoperative chemotherapy
19/91 (20.9)
13/141 (9.2)
Age C 60 years
Malignant diagnosis
2/50 (4)
[0.2
10/66 (15.2) [0.2
9/67 (13.4)
[0.2
[0.2
16/96 (16.7) 5/28 (17.9)
[0.2
20/135 (14.8) [0.2a
NS
14/79 (17.7)
[0.2a
NS
10/80 (12.5) [0.2
18/153 (11.8)
2/10 (20)
[0.2
23/153 (15)
BMI body mass index, NS not significant a
Included in the multivariate model
proportion of bundle patients received this antibiotic compared with the pre-bundle group. The relative contribution of the individual elements of the bundle to the improved outcome is unclear. Some authors recommend cefazolin as the preoperative antibiotic for hepatic surgery.5,23 The choice of ampicillin/sulbactam as the preoperative antibiotic in our bundle was based on our observations, prior to bundle initiation, that many organisms growing from postoperative abscess cultures were resistant to cefazolin. However, at the conclusion of the study we noted that patients commonly grew organisms from abscesses that were resistant to their respective preoperative antibiotic (Table 5). Therefore, we feel it is unlikely that the use of ampicillin/sulbactam was a major factor in the decreased postoperative infection rate. It is more likely that the other elements of the bundle contributed, to a greater degree, to the decrease in infection rate. Chlorhexadine showers were added to the bundle because it has been established that skin bacteria counts are decreased with chlorhexadine application.5 Although a
meta-analysis questions the effectiveness of preoperative bathing with antiseptic solutions,24 a clinical trial demonstrated decreased SSIs with chlorhexadine bathing.25 We chose chlorhexadine, rather than povidone-iodine, for the skin preparation because randomized, prospective studies showed decreased local positive culture results and decreased SSIs after surgery.5,15 We included IobanÒ drapes in the bundle because a clinical trial evaluating the use of an iodophor impregnated skin drape in hepatic surgery demonstrated a lower SSI rate with use.26 Standardized antibiotic redosing was implemented based on the specific drug half-life because it has been shown that therapeutic drug levels at the time of wound closure are critical determinants of SSI rates.18 There is extensive data on the use of a topical antimicrobial in preventing SSIs. Clinical trials have shown a decrease in SSI rates in both biliary surgery as well as surgery localized to the subcutaneous tissue 19–21 The decision was therefore made to add the broad-spectrum neomycin–polymixin–bacitracin wound irrigation to our bundle.
M. V. Hill et al. TABLE 5 Antibiogram: culture data from deep SSI and sensitivities to preoperative antibiotics Prophylactic cephalexin
Prophylactic ampicillin/ sulbactam
Gram-positive organisms Corynebacterium
NA
Coagulase-negative staphylococci 1
R
2
R
MRSA
R
Propionibacterium acnes 1
NA
2 Streptococci viridans
S S
Enterococcus 1
R
2 3 4
S S R
Gram-negative organisms Serratia
R
Escherichia coli 1
S
2
S
CONCLUSIONS
3
S
4
S
Pseudomonas 1
S
2 Citrobacter
R R
Enterobacter 1
R R
2
R
3
R
Morganella morganii
R
Anaerobes Bacteroides fragilis
S
Clostridium perfringens Any resistant organisms
We were able to show that a low-cost, easy to implement bundle was able to decrease not only postoperative infection rates but also total complications, LOS, and cost. This suggests that such bundles should be routinely developed and implemented in order to optimize outcomes in this at-risk population.
R
2 Klebsiella 1
preparation.7 The colorectal bundle of the Mayo Clinic, Rochester, included chlorhexadine preoperative washes, chlorhexadine preparation, and routine antibiotic redosing.8 This intervention study, like other studies of infection prevention bundles,7,8 lacks the rigor of a randomized, prospective trial in determining the value of an infection prevention bundle. However, the strengths of the study include the uniformity of the patients treated and the surgical techniques used pre- and post-bundle; all patients were treated by a single surgeon, at a single institution. Uniform compliance with the bundle interventions was another study strength.27 Other investigators have reported that diabetes, obesity, and excessive blood loss are risk factors for postoperative infections in liver surgery patients.10,12,13 In our multivariate analysis, we also found that diabetes was an independent predictor of post-operative infections. Although our diabetic patients underwent routine blood glucose checks and were treated with sliding-scale insulin, we did not institute a strict regimen for glucose control. Such regimens have recently been shown to be effective in reducing SSIs in diabetic patients undergoing surgery, and will be considered as an addition to our bundle in the future.28
NA 8/12 (66.7 %)
6/12 (50 %)
Numbers represent separate isolates of bacteria R resistant, S sensitive, NA no sensitivities reported, MRSA methicillin-resistant Staphylococcus aureus
The content of our bundle shared some elements utilized in other infection prevention bundles. The pancreaticoduodenectomy bundle created by Lavu et al. contained preoperative chlorhexadine wipes and chlorhexadine
REFERENCES 1. Coello R, Charlett A, Wilson J, Ward V, Pearson A, Boriello P. Adverse impact of surgical site infections in English hospital. J Hosp Infect. 2005;60:93–103. 2. Forbes S, Stephen W, Harper W, Loeb M, Smith R, Christofferson E, et al. Implementation of evidence-based practices for surgical site infection prophylaxis: results of a pre- and postintervention study. J Am Coll Surg. 2008;207:336–341. 3. Astagneau P, Rious C, Golliot F, Brucker G. Morbidity and mortality associated with surgical site infections: results from the 1997-1999 INCISO surveillance. J Hosp Infect. 2001;48:267–274. 4. Mangram AJ, Horan TC, Pearson ML, Silver LC, Jarvis WR. Guideline for prevention of surgical site infection, 1999. Am J Infect Control. 1999;27:97–134. 5. Alexander JW, Solomkin J, Edwards M. Updated recommendations for control of surgical site infections. Ann Surg. 2011;253:1082–1093. 6. de Lissovoy G, Fraeman K, Hutchins V, Murphy D, Song D, Vaughn B. Surgical site infection: incidence and impact of hospital utilization and treatment costs. Am J Infect Control. 2009;37:387–397.
Perioperative Bundle Decreases Postoperative Hepatic Surgery Infections 7. Lavu H, Klinge M, Nowcid L, Cohn H, Grenda D, Sauter P, et al. Perioperative surgical care bundle reduces pancreaticoduodenectomy wound infections. J Surg Res. 2012;174:215–221. 8. Cima R, Dankbar E, Lovely J, Pendlimari R, Aronhalt K, Nehring S, et al. Colorectal surgery surgical site infection reduction program: a national surgical quality improvement program-driven multidisciplinary single-institution experience. J Am Coll Surg. 2013;216:23–33. 9. Elola-Olaso AM, Davenport DL, Hundley JC, Daily MF, Gedaly R. Predictors of surgical site infection after liver resection: a multicentre analysis using national surgical quality improvement program data. HPB (Oxford). 2012;14:136–141. 10. Okabayashi T, Nishimori I, Yamashita K, Sugimoto T, Yatabe T, Maeda H, et al. Risk factors and predictors for surgical site infection after hepatic resection. J Hosp Infect. 2009;73:47–53. 11. Ceppa E, Pitt H, House M, Kilbane M, Nakeeb A, Schmidt C, et al. Reducing surgical site infections in hepatopancreatobiliary surgery. HPB (Oxford). 2013;15:384–391. 12. Gedaly R, McHugh P, Johnston T, Jeon H, Ranjan D, Davenport D. Obesity, diabetes, and smoking are important determinants of resource utilization in liver resection: a multicenter analysis of 1029 patients. Ann Surg. 2009;249(3):414–419. 13. Kobayashi S, Gotohda N, Nakagohri T, Takahashi S, Konishi M, Kinoshita T. Risk factors of surgical site infection after hepatectomy for liver cancers. World J Surg. 2009;33:312–317. 14. Moreno Elola-Olaso A, Davenport D, Hundley J, Daily M, Gedaly R. Predictors of surgical site infection after liver resection: a multicentre analysis using national surgical quality improvement program data. HPB (Oxford). 2012;14:135–141. 15. Pessaux P, van den Broek M, Wu T, Olde Damink S, Dejong C, Ntourakis D, et al. Identification and validation of risk factors for postoperative infectious complications following hepatectomy. J Gastrointest Surg. 2013;17:1907–1916. 16. Bibbo C, Patel D, Gehrmann R, Lin S. Chlorhexadine provides superior skin decontamination in foot and ankle surgery. Clin Orthop Relat Res. 2005;438:204–208. 17. Hawn M, Itani K, Gray S, Vick C, Henderson W, Houston T. Association of timely administration of prophylactic antibiotics for major surgical procedures and surgical site infection. J Am Coll Surg. 2008;206:815–821.
18. Zelenitsky S, Ariano R, Harding G, Silverman R. Antibiotic pharmacodynamics in surgical prophylaxis: an association between intraoperative antibiotic concentrations and efficacy. Antimicrob Agents Chemother. 2002;46(9):3026–3030. 19. Lord JW, LaRaja RD, Daliana M, Gordon MT. Prophylactic antibiotic wound irrigation in gastric, biliary and colonic surgery. Am J Surg. 1983;145(2):209–212. 20. Lord J, Rossi G, Daliana M. Intraoperative antibiotic wound lavage. Ann Surg. 1977;185(6):634–641. 21. Halasz NA. Wound infection and topical antibiotics. Arch Surg. 1977;112:1240–1244. 22. Reeves JG, Suriawinata AA, Ng DP, Holubar SD, Mills JB, Barth RJ. Short-term preoperative diet modification reduces steatosis and blood loss in patients undergoing liver resection. Surgery. 2013;154(11):1031–1037. 23. Bratzler D, Houck P. Antimicrobial prophylaxis for surgery: an advisory statement from the national surgical infection prevention project. Am J Surg. 2005;189:395–404. 24. Webster J, Osborne S. Preoperative bathing or showering with skin antiseptics to prevent surgical site infection. Cochrane Database Syst Rev. 2012;9:CD004985. 25. Wihlborg O. The effect of washing with chlorhexadine soap on wound infection rate in general surgery. A controlled clinical study. Ann Chir Gynaecol. 1987;76(5):263–5. 26. Yoshimura Y, Kubo S, Hirohashi K, Ogawa M. Plastic iodophor drape during liver surgery operative use of the iodophor-impregnated adhesive drape to prevent wound infection during high risk surgery. World J Surg. 2003;27:685–8. 27. Waits SA, Fritze D, Banerjee M, Zhang W, Kubus J, Englesbe MJ, et al. Developing an argument for bundled interventions to reduce surgical site infection in colorectal surgery. Surgery. 2014;115:602–606. 28. Wallaert JB, Nolan BW, Adams J, Stanley A, Eldrup-Jorgensen J, Cronenwett JL, et al. The impact of diabetes on postoperative outcomes following lower-extremity bypass surgery. J Vasc Surg. 2012;56:1317–1323.
