Reducing surgical site infections through a multidisciplinary computerized process for preoperative prophylactic antibiotic administration

The American Journal of Surgery 192 (2006) 663– 668

Presentation

Reducing surgical site infections through a multidisciplinary computerized process for preoperative prophylactic antibiotic administration Alexandra L.B. Webb, M.D.a,b,*, Rene L. Flagg, R.N.a, Aaron S. Fink, M.D.a,b a

Department of Surgery, Atlanta Veterans Affairs Medical Center, Decatur, GA, USA b Department of Surgery, Emory University School of Medicine, Atlanta, GA, USA Manuscript received June 23, 2006; revised manuscript August 3, 2006

Presented at the 30th Annual Surgical Symposium of the Association of VA Surgeons, Cincinnati, Ohio, May 7–9, 2006

Abstract Background: Surgical site infections (SSIs) result in significant postoperative morbidity and mortality. Although many of these infections can be prevented by timely administration of preoperative antibiotics, data suggest that many patients do not receive such therapy. Methods: A multidisciplinary team was convened that reviewed published guidelines, made antibiotic recommendations, and addressed administration issues. Responsibility for antibiotic administration was shifted from preoperative nursing staff to the anesthetist. Electronic quick orders were developed to encourage appropriate antibiotic selection and simplify order creation. Results: Timely administration of preoperative antibiotics improved from 51% to 98% from February 2005 to February 2006. Appropriate antibiotic administered improved from 78% to 94%. The clean wound infection rate decreased from 2.7% to 1.4% over the same time period. Conclusion: A multidisciplinary approach to prophylactic antibiotic use, including computer-guided decision support, facilitates appropriate preoperative antibiotic use, resulting in a significant decrease in surgical wound infections. © 2006 Excerpta Medica Inc. All rights reserved. Keywords: Surgical infection prophylaxis; Computer-guided decision support; Wound infection

Surgical site infections (SSIs) are an important cause of morbidity, mortality, and increased costs in today’s healthcare environment. These events represent the most common nosocomial infection in surgical patients [1]. The overall incidence of SSIs has been reported to be 2% to 5% in patients undergoing clean extra-abdominal procedures and up to 20% for intra-abdominal procedures [2]. Development of an SSI has been associated with an increase in length of hospital stay of 4 to 7 days and an incremental increase in hospital costs by as much as $3,000 [3,4]. Many factors have been identified which may reduce the incidence of SSIs, including method of preoperative skin preparation [5,6], use of antimicrobial showers [7], method/ timing of hair removal [8,9], maintenance of perioperative normoglycemia [10 –12], and normothermia [13,14]. * Corresponding author. Atlanta VA Medical Center, Department of Surgery, 1670 Clairmont Rd. (112), Decatur, GA 30033. Tel.: ⫹1-404728-7621; fax: ⫹1-404-329-2201. E-mail address: [email protected]

Preoperative antibiotic prophylaxis has repeatedly been shown to be effective in reducing SSI [15–17]. Administration of antibiotics within 1 hour prior to surgical incision (2 hours for vancomycin) has been associated with lower infection rates [18]. It is also important to re-dose antimicrobial agents, particularly cephalosporins, during lengthy procedures. Despite an advanced electronic medical record and electronic order entry system (CPRS) [19], it was noted that patients at our institution were not receiving appropriate preoperative antibiotics in a timely fashion. In the following sections, we describe the interventions made to address this problem. Methods A multidisciplinary team was assembled, including representatives from surgery, anesthesia, infectious disease, nursing, and pharmacy. This team reviewed the process of antibiotic ordering and administration and identified problem areas in this process. Several barriers to timely antibi-

0002-9610/06/$ – see front matter © 2006 Excerpta Medica Inc. All rights reserved. doi:10.1016/j.amjsurg.2006.08.014

664

A.L.B. Webb et al. / The American Journal of Surgery 192 (2006) 663– 668

Table 1 Local antibiotic recommendations Service

Procedure

Primary choice

PCN allergy

Cardiac Thoracic General

All All Cholecystectomy (elective) Breast Breast (high risk) Hernia Esophagus/stomach/small intestine/hepatectomy

Vancomycin Cefuroxime None None Cefazolin Cefazolin Cefazolin Cefuroxime Cefotetan Cefotetan Cefazolin

Vancomycin Vancomycin

Gynecology ENT Oral surgery

Orthopedics GU

Vascular Plastics Podiatry

Hepatobiliary (acute cholecystitis/indwelling drains) Appendix/colorectal/anal All Clean Clean-contaminated Clean Implants, open fractures, orthonathics Bone graft Clean, no implants Clean, with implants (ORIF/spine/total joints) Low-risk High-risk All other procedures DM foot Clean

None Cefazolin None Penicillin Cefazolin None Cefazolin None Ciprofloxacin Ampicillin/gentamicin Cefazolin Ampicillin/sulbactam Cefazolin

Clean cases DM foot

Cefazolin Ampicillin/sulbactam

Clindamycin Vancomycin Vancomycin Clindamycin/ciprofloxacin Clindamycin/ciprofloxacin Clindamycin Doxycycline Clindamycin Clindamycin Clindamycin Vancomycin Ciprofloxacin Vancomycin Clindamycin/ciprofloxacin Clindamycin Vancomycin Clindamycin Clindamycin/ciprofloxacin

PCN ⫽ penicillin; DM ⫽ diabetic; ORIF ⫽ open reduction internal fixation.

otic administration were identified. It was noted that a failure at any one of many steps in this complex process could prevent meeting these goals. Problems identified included inappropriate antibiotic ordered, antibiotic given too early, antibiotic not ordered, and antibiotic not immediately available for administration. To address the problem of inappropriate antibiotic choice, a clear definition of appropriate antibiotics was needed. The panel recommended prophylactic surgical antibiotics based on a review of various national guidelines [1,2,20 –22], local antibiograms, and local cost data (Table 1). These recommendations were then reviewed with each surgical section chief for their comment and approval. This list of recommended prophylactic antibiotics, categorized by subspecialty and type of procedure, was then adopted as the standard for this institution. Next, the issue of timeliness was examined. Previously, antibiotics were ordered “on call to operating room.” Upon further discussion, it became clear that this was not sufficiently explicit: patients were given antibiotics on the inpatient ward prior to transport to the preoperative area or in the preoperative area as long as 3 to 5 hours prior to surgery. It was difficult to predict when a procedure would begin, given the unknown time for transport to the operating room, induction of anesthesia, positioning, prepping, and draping. Thus, we opted to shift the responsibility for antibiotic administration to Anesthesia in order to improve timeliness. The panel determined that the issue of lack of antibiotic availability was closely linked to failure to order prophy-

lactic antibiotics. On our preoperative order form, the section for prophylactic antibiotic orders was easily overlooked (Fig. 1). In addition, the lack of focused guidelines led to diverse antibiotic choices, complicating pharmacy’s ability to deliver antibiotics in a timely fashion, thereby ensuring timely availability of the appropriate antibiotic. Solutions to each of these problems were proposed and subsequently implemented in a stepwise fashion between February and July of 2005. First, the process changes were implemented. Next, electronic orders were rolled out to each surgical subspecialty consecutively in order to ensure a smooth transition and address any problems as that arose. To measure the clinical impact of these changes, most surgical patients (⬃90%) were monitored monthly for: (1) percentage of patients receiving appropriate preoperative antibiotics; (2) percentage receiving preoperative antibiotics within 1 hour prior to surgical incision (2 hours for vancomycin); and (3) appropriateness of antibiotics as determined by adherence to the antibiotics recommended by the multidisciplinary panel. In addition, the clean wound infection rate as reported by the National Surgical Quality Improvement Program (NSQIP) [23] was recorded and correlated with our antibiotic administration data. Ophthalmology, dialysis access, and cystoscopy patients were excluded from review. Results Multiple changes were implemented to address the challenges described above. First, as described above, a list of

A.L.B. Webb et al. / The American Journal of Surgery 192 (2006) 663– 668

Fig. 1. Paper preoperative order form.

appropriate antibiotics was generated to establish a consensus for this institution (Table 1). This consensus was then used to generate electronic quick orders, thereby facilitating computer-enhanced decision making by providers. To address the problem of early antibiotic administration, a clear administration protocol was established. The nursing staff in the preoperative area was tasked to prepare and hang the antibiotic, but not to initiate infusion. In turn,

665

the anesthesia staff was instructed to infuse the medication and record the time of infusion upon the patient’s arrival in the operating room. This new process ensured timely administration without adding undue burden to the anesthesia staff at an already busy time. Electronic quick orders were developed to address the issues of inappropriate antibiotic choice and failure to order an antibiotic in advance. A preoperative order page was developed in CPRS, which facilitated the preoperative assessment and preparation of the patient (Fig. 2). It included sections for patient status (same day admission, 23-hour observation, or outpatient), preoperative laboratory studies, x-rays, electrocardiogram, blood products, deep venous thrombosis prophylaxis, bowel preparation, and prophylactic antibiotics. This served to remind the provider to address all of these issues for each patient. This order structure served to ensure that antibiotics are ordered in advance and are therefore ready for administration at the time of surgery. In addition, in most instances, two antibiotic choices are given for each type of procedure, one preferred choice and a second choice for patients with a beta-lactam allergy (Fig. 3). Such computer guidance in decision making serves to educate providers while ensuring optimal patient care. In addition, decreasing the number of different antibiotics given preoperatively allowed development of focused pharmacy protocols, facilitating immediate availability of these antibiotics in the preoperative area. Examination of the clinical impact of these changes yielded dramatic results: Fig. 4 shows the percentage of antibiotics administered in a timely fashion on a monthly basis. It should be noted that prior to the above interventions, only 51% of antibiotics were administered timely and this subsequently improved to greater than 95%. Fig. 5 demonstrates percentage of appropriate antibiotics administered. A substantial improvement was noted from January to May 2005; frequency of appropriate administration has consistently remained above 90% since that time.

Fig. 2. Electronic preoperative orders.

666

A.L.B. Webb et al. / The American Journal of Surgery 192 (2006) 663– 668

Fig. 3. Electronic quick orders for antibiotics.

Fig. 6 follows the clean wound infection rate over this time period. A marked decrease in this rate was noted from February 2005 to August 2005. Comments Previous studies have clearly demonstrated that preoperative antibiotic prophylaxis is rarely administered appropriately, despite the strong evidence supporting this practice. In 1996, Silver et al studied patients undergoing abdominal aortic aneurysm repair, hip replacement, or colectomy in New York State [24]. They found that 14% received no prophylactic antibiotics and only 63% received antibiotics in the 2 hours prior to surgical incision. In 2005, the National Surgical Infection Prevention Project evaluated 34,000 Medicare patients undergoing cardiac, vascular, and colorectal procedures, as well as hip or knee replacement and hysterectomy. They found that only 56% of patients had an antimicrobial agent administered within appropriate time frames [25]. Surgical infection prophylaxis (SIP) is receiving increasingly public attention. By way of example, guidelines for SSI prevention have recently been published by the Centers for Disease Control, Centers for Medicare and Medicaid Services, Joint Commission on Accreditation for Healthcare

100%

Organizations, and the Department of Veterans Affairs. These guidelines are currently being used as performance measures, and may ultimately be used in developing payfor-performance initiatives. Hopefully, such system-based guidelines will raise awareness for the need to tailor antibiotics to the expected organisms for a given procedure. Previously, it has been demonstrated that Veterans Affairs (VA) hospitals have failed to meet the standards established by these guidelines [26,27]. As a result, CMS introduced and the VA adopted 3 performance measures in fiscal year 2005: SIP-1, SIP-2, and SIP-3. These performance measures are now being used as a measure of quality of care and a method by which to compare performance among VA facilities. SIP-1 deals with timeliness of preoperative antibiotic administration, SIP-2 with appropriateness, and SIP-3 is concerned with timely discontinuation of prophylactic antibiotics. The VA has been a pioneer in the development and implementation of the electronic medical record. Currently, the VA boasts a highly integrated system, encompassing electronic order entry, electronic documentation and signature, and digital image viewing, all within CPRS. Some evidence suggests that the use of computer-guided decision making can enhance performance and adherence to antibiotic guidelines [28,29]. Our results further support this the-

100%

90%

90%

80%

80%

70% 60%

70%

50%

60%

40%

Feb Mar Apr May Jun 05 05 05 05 05

Jul 05

Aug Sep Oct Nov Dec Jan Feb 05 05 05 05 05 06 06

Fig. 4. Percentage of antibiotics administered in a timely manner.

Ja n0 Fe 5 b0 M 5 ar -0 Ap 5 r- 0 M 5 ay -0 Ju 5 n05 Ju l-0 Au 5 g0 Se 5 p0 O 5 ct -0 No 5 v0 De 5 c05 Ja n0 Fe 6 b06

50%

30%

Fig. 5. Percentage of patients given appropriate prophylactic antibiotics.

A.L.B. Webb et al. / The American Journal of Surgery 192 (2006) 663– 668

667

that improved timeliness and appropriateness of antibiotic prophylaxis merely delayed infection in these patients. Finally, re-dosing of antibiotics was not examined in this study, nor was prompt discontinuation of antibiotics, rendering comment on the impact of these measures impossible.

3.50% 3.00% 2.50% 2.00% 1.50% 1.00% 0.50% 0.00% Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar 05 05 05 05 05 05 05 05 05 05 05 06 06 06

Fig. 6. Clean wound infection rate.

ory. Computer-guided decision support is particularly useful in a teaching institution, where there is monthly resident turnover, as well as varying levels of experience among providers ordering prophylactic antibiotics. Several factors contributed to the success of this intervention. The input of all groups involved in the process of ordering and administering antibiotics was essential to identify possible pitfalls to any proposed solution. In addition, the “buy-in” of surgical section chiefs had a trickle down effect, encouraging compliance with recommended antibiotic choices among ordering providers. Ideally, all patients should receive appropriate prophylactic antibiotics in a timely fashion. Although our efforts have been most successful, we have faced some challenges in attaining this goal. One obvious requirement is that the desired antibiotic be available for administration. Recently, the manufacturer of cefotetan discontinued its production. Many of the antibiotics that did not meet appropriateness criteria were ordered in an effort to find a substitute for cefotetan. Also, timely administration of preoperative antibiotics for inpatients continues to be a challenge. We continue to examine the process for ordering and delivering antibiotics to these patients to improve our performance in this arena. Given the increased focus of surgeons, nursing staff, and anesthesia on the importance of antibiotic prophylaxis, a seemingly paradoxical challenge arises: that of patients receiving prophylactic antibiotics when none are indicated. Our electronic quick orders include a statement that no antibiotic is indicated for clean procedures involving the skin and soft tissues, where no foreign material is implanted. However, it has proven difficult for surgeons to defer requests for prophylactic antibiotics in these patients. While our results associate a decrease in the clean wound infection rate with improved adherence to antibiotic prophylaxis guidelines, this study has several limitations: First, wound infection rates for clean-contaminated procedures were not evaluated. However, since a benefit is demonstrated in patients with clean wounds, where infection is much less frequent, one could infer that such a benefit is likely to apply to contaminated procedures as well. Also, infections in clean wounds presenting more than 30 days postoperatively were not detected in this study, since the NSQIP data do not extend beyond this time. This is the case for both pre- and post-intervention groups, mitigating the impact of this limitation. We cannot rule out the possibility

Conclusions A multidisciplinary approach combined with computerguided decision support is successful in improving compliance with guidelines for antibiotic prophylaxis in surgery. This improved compliance is correlated with a decrease in the clean wound infection rate. Acknowledgment The authors would like to the following individuals for their contributions to this project: Robert Gaynes, M.D., Department of Medicine/Infectious Disease, John Scharf, M.D., Department of Anesthesia, Catherine Connell, Pharm.D., and Cheryl Frank, R.N., all of the Atlanta Veterans Affairs Medical Center. References [1] Mangram AJ, Horan TC, Pearson ML, et al. Guideline for prevention of surgical site infection, 1999. Infect Control Hosp Epidemiol 1999; 20:247– 80. [2] Auerbach AD. Prevention of surgical site infections. In: Shojania KG, Duncan BW, McDonald KM, et al, editors. Making Health Care Safer: A Critical Analysis of Patient Safety Practices. Evidence report/technology assessment no. 43. AHRQ publication no 01-E058. Rockville, MD: Agency for Healthcare Research and Quality; 20 July 2001:221– 44. Available at: http://www.ahrq.gov/clinic/ptsafety/pdf/ ptsafety.pdf. Accessed 1 June 2006. [3] Martone WJ, Jarvis WR, Culver DH, et al. Incidence and nature of endemic and epidemic nosocomial infections. In: Bennett JV, Brachman PS, editors. Hospital Infections. 3rd ed. Boston: Little Brown; 1992:577–96. [4] Dimick JB, Chen SL, Taheri PA, et al. Hospital costs associated with surgical complications: a report from the private-sector National Surgical Quality Improvement Program. J Am Coll Surg 2004;199: 531–7. [5] Larson E. Guideline for use of topical antimicrobial agents. Am J Infect Control 1988;16:253– 66. [6] Aly R, Maibach HI. Comparative antibacterial efficacy of a 2-minute surgical scrub with chlorhexidine gluconate, povidone-iodine, and chloroxylenol sponge-brushes. Am J Infect Control 1988;16:173–7. [7] Garibaldi RA. Prevention of intraoperative wound contamination with chlorhexidine shower and scrub. J Hosp Infect 1988;11(suppl B):5–9. [8] Alexander JW, Fischer JE, Boyajian M, et al. The influence of hair-removal methods on wound infections. Arch Surg 1983;118: 347–52. [9] Masterson TM, Rodeheaver GT, Morgan RF, Edlich RF. Bacteriologic evaluation of electric clippers for surgical hair removal. Am J Surg 1984;148:301–302. [10] Zerr KJ, Furnary AP, Grunkemeier GL, et al. Glucose control lowers the risk of wound infection in diabetics after open heart operations. Ann Thorac Surg 1997;63:356 – 61. [11] Furnary AP, Kerr KJ, Grunkemeier GL, et al. Continuous intravenous insulin infusion reduces the incidence of deep sternal wound infection in diabetic patients after cardiac surgical procedures. Ann Thorac Surg 1999;67:352– 60. [12] Latham R, Lancaster AD, Covington JF, et al. The association of diabetes and glucose control with surgical-site infections among cardiothoracic surgery patients. Infect Control Hosp Epidemiol 2001;22: 607–12. [13] Kurz A, Sessler DI, Lenhardt RA. Perioperative normothermia to reduce the incidence of surgical-wound infection and shorten hospitalization. N Engl J Med 1996;334:1209 –15.

668

A.L.B. Webb et al. / The American Journal of Surgery 192 (2006) 663– 668

[14] Melling AC, Ali B, Scott EM, et al. Effects of preoperative warming on the incidence of wound infection after clean surgery: a randomized controlled trial. Lancet 2001;358:876 – 80. [15] Burke JF. The effective period of preventive antibiotic action in experimental incisions and dermal lesions. Surgery 1961;50:161– 8. [16] Stone HH, Hooper CA, Kolb LD, et al. Antibiotic prophylaxis in gastric, biliary, and colonic surgery. Ann Surg 1976;184:443–52. [17] Polk HC Jr, Trachtenberg L, Finn MP. Antibiotic activity in surgical incisions: the basis for prophylaxis in selected operations. JAMA 1980;244:1353– 4. [18] Classen DC, Evans RS, Pestotnik SL, et al. The timing of prophylactic administration of antibiotics and the risks of surgical-wound infection. N Engl J Med 1992;326:281– 6. [19] Meldrum K, Volpp B, Vertigan R. Deparment of Veterans Affairs’ computerized patient record system. Proc AMIA Symp 1999;12:1214. [20] American Society of Health-System Pharmacists. ASHP therapeutic guidelines on antimicrobial prophylaxis in surgery. Am J Health Syst Pharm 1999;56:1839 – 88. [21] Scottish Intercollegiate Guidelines Network (SIGN). Antibiotic Prophylaxis in Surgery. A National Clinical Guideline. SIGN publication no. 45. SIGN; 2000:1–36. [22] Bratzler DM, Houck PM. Antimicrobial prophylaxis for surgery: an advisory statement from the National Surgical Infection Prevention Project. CID 2004;38:1706 –15.

[23] Khuri SF, Daley J, Henderson W, et al. The Department of Veterans Affairs’ NSQIP: the first national, validated, outcome-based, riskadjusted, and peer-controlled program for the measurement and enhancement of the quality of surgical care. National VA Surgical Quality Improvement Program. Ann Surg 1998;228:491–507. [24] Silver A, Eichorn A, Kral J, et al. Timeliness and use of antibiotic prophylaxis in selected inpatient surgical procedures. Am J Surg 1996;171:548 –52. [25] Bratzler DW, Houck PM, Richards C, et al. Use of antimicrobial prophylaxis for major surgery: Baseline results from the National Surgical Infection Prevention Project. Arch Surg 2005;140:174 – 82. [26] Munster AM, Weiner J, Gibson G. Prophylactic antibiotics in surgery. Practices within surgical services of the Veterans Administration. JAMA 1979;241:717– 8. [27] Matuschka PR, Cheadle WG, Burke JD, et al. A new standard of care: administration of preoperative antibiotics in the operating room. Am Surg 1997;63:500 –3. [28] Pestotnik SL, Classen DC, Evans RS, et al. Implementing antibiotic practice guidelines through computer-assisted decision support: clinical and financial outcomes. Ann Intern Med 1996;124:884 –90. [29] Evans RS, Pestotnik SL, Classen DC, et al. A computer-assisted management program for antibiotics and other antiinfective agents. N Engl J Med 1998;338:232– 8.