Prepping for Prevention: An Intervention to Optimize Skin Antisepsis and Peri-Operative Antibiotic Prophylaxis

Abstract

Background:

The U.S. Centers for Disease Control and Prevention recommend bathing prior to surgery, surgical skin antisepsis, peri-operative antibiotic administration, normothermia throughout the procedure, serum glucose concentration <200 mg/dL throughout the procedure, and hyperoxygenation in the immediate post-operative period to prevent surgical site infection (SSI). We developed interventions to standardize skin antisepsis and peri-operative antibiotic administration at our institution.

Methods:

This is a cross-sectional evaluation of surgical skin antisepsis and antibiotic administration before and after a series of interventions designed to standardize the processes.

Results:

One hundred twenty-four surgical skin antisepsis opportunities were observed; significant improvement was seen in hand hygiene prior to performing skin antisepsis (compliance changing from 1% to 48%; p < 0.001), sleeves being worn during skin antisepsis (1% versus 67%; p < 0.001), use of the correct cleansing time (47% versus 85%; p < 0.001), allowance for adequate drying time (67% versus 87%; p = 0.02), and use of a cleansing motion from the incision to the periphery (78% versus 95%; p = 0.004). Pre-operative antibiotic order placement, correct antibiotic selection, and optimal antibiotic dose were evaluated in 466 surgical procedures. Significant improvement was seen in both peri-operative order placement (59% versus 70%; p = 0.02) and correct antibiotic selection (52% versus 95%; p < 0.001).

Conclusion:

An intervention to standardize skin antisepsis and to encourage early ordering of peri-operative antibiotics was successful.

Surgical site infections (SSIs) are the most common type of healthcare-associated infections [1], affecting 500,000 patients annually or almost 3% of those who have recently undergone surgery [2]. Such infections are associated with an increase in the length of stay by 11 days, a higher re-admission rate, and a two- to 11-fold increase in the risk of death [3]. The financial burden to the healthcare system in the United States is estimated to be $2.8–$3.8 billion annually as measured in 2012 U.S. dollars [3].

More than half of SSIs are preventable [4]. Quality improvement interventions that prevent SSIs can save lives, improve patients' quality of life, and reduce medical expenses. The U.S. Centers for Disease Control and Prevention (CDC) have published guidelines on the prevention of SSIs, outlining six essential practices. These are bathing prior to surgery, surgical skin antisepsis, peri-operative antibiotic administration, maintenance of normothermia throughout the surgical procedure, control of the serum glucose concentration <200 mg/dL throughout the procedure, and hyperoxygenation in the immediate post-operative period [5]. Of these six essential practices, optimal peri-operative antibiotic administration and a high-quality application of surgical skin antisepsis (commonly referred to as the “skin prep”) have remained universally recommended since 1999 [5,6].

We performed an audit of skin preparation practices and peri-operative antibiotic administrations at our institution. Using the Plan, Do, Study, Act (PDSA) method, we developed and implemented a quality improvement intervention to address deficiencies and improve adherence to best practice recommendations.

Patients and Methods

Setting and population

Denver Health Medical Center is a 525-bed, academic, safety-net, Level I trauma hospital that performs more than 17,000 operations annually [7]. There are 16 operating rooms and 45 circulating nurses. In addition to staff surgeons and circulating nurses, care for patients is provided by surgical residents, nurse practitioners, physician assistants, and temporary, traveling nurses. The study period was between September 1, 2016, and February 28, 2018. This study was a cross-sectional evaluation of skin preparation and antibiotic administration in two time periods.

Circulating nurses who completed a skin preparation for any surgical procedure were eligible for inclusion in these evaluations. Although inclusion in the study was voluntary, none declined to participate. Cases eligible for inclusion in the peri-operative antibiotic prophylaxis were colorectal operations, abdominal hysterectomies, breast surgery, hip and knee arthroplasties, gastric procedures, craniotomies, spinal fusions, vaginal hysterectomies, and open reduction and internal fixation of fractures. These ten procedure types were chosen because there were recent and historic SSI data available. Patients were excluded if they met any of the following criteria: under the age 18 years, had undergone emergency surgery, and had existing infections at admission. Additionally, cases were excluded from the peri-operative antibiotic prophylaxis evaluation if the patient died before the SSI surveillance period was completed (30 or 90 days; based on National Healthcare Safety Network (NHSN) criteria) [8].

Intervention

Skin preparation application

To improve the quality of the surgical skin preparation practices, a multidisciplinary committee was formed consisting of circulating nurses, scrub technicians, infection prevention staff, and the nurse educator for the peri-operative department. The committee members became subject matter experts in skin preparation best practices by studying the Association of periOperative Registered Nurses (AORN) guidelines and the manufacturer's recommendations for the use of the antiseptic products. Additionally, the antiseptic manufacturer provided the committee with a sales representative to provide further clarity and guidance in the use of the antiseptics. The committee members determined that appropriate skin preparation would be adjudicated on six criteria: Hand hygiene before the preparation, sleeves worn, gloves worn, adequate cleansing time given, adequate drying time allowed, and a cleansing motion that moved from the incision site to the periphery. The former three criteria are defined by AORN [9], whereas the latter three are defined by the manufacturer's indications for product use. Specifically, when performing a surgical preparation with 4% chlorhexidine soap (Hibiclens), the agent is applied liberally to the surgical site and swabbed for at least two minutes. The preparation begins at the incision site and moves to the periphery. The prepared area is blotted dry with a sterile towel and the entire process repeated for an additional two minutes [10]. When performing a skin preparation with 2% chlorhexidine with 70% isopropyl alcohol (ChloraPrep), the skin preparation begins at the incision site using a back-and-forth motion for 30 seconds. After 30 seconds, the same back and forth motion is used moving to the periphery. This is continued for two additional minutes. At the conclusion of 2.5 minutes, the product is allowed to dry for at least three minutes [10]. A standardized audit form was developed, and the manufacturer's representatives performed an audit to characterize our institution's adherence to best practice recommendations.

Using the baseline audit data, the committee utilized the PDSA method and determined that there were deficiencies in education among circulating nurses and decided to utilize peer-to-peer teaching. First, the committee determined that they needed hands-on simulated practice before they could coach their coworkers adequately. During a meeting, each committee member was given two 3 × 5-inch index cards that detailed evidence-based guidelines for the two most widely used antiseptics: ChloraPrep and Hibiclens. With a mannequin and all necessary supplies, committee members role-played teaching each other proper skin preparation technique. Committee members asked and answered technical questions by consulting the AORN guidelines and the antiseptic sales representative.

A list of all circulating nurses was then posted near the operating rooms. In August 2017, committee members conducted peer-to-peer teaching with their colleagues throughout the workday and ensured the skin preparation competence of each nurse. Additionally, a 45-minute in-service class for staff members was offered during which the antiseptic products' nurse educators taught the staff about the recommended usage of the products, and the committee members spoke about the AORN guidelines. After the initial round of education was completed, the second audit was performed.

Again using the PDSA method, the committee identified equipment barriers as well as continued educational deficiencies. To adhere to AORN guidelines, the nurses must perform hand hygiene, don gloves, and wear sleeves during the skin preparation. However, there was a paucity of hand hygiene stations in the operating rooms. Furthermore, there were insufficient sterile sleeves for the volume of skin preparations performed throughout the day. In the short term, waterless hand sanitizer was added to each nurse's desk, and the quantity of sterile sleeves stocked near the operating rooms was tripled. The nurses also were provided with nonsterile, disposable jackets to use as an alternative to sleeves. However, the committee was concerned about long-term adherence to these best practice recommendations. As a result, the committee members designed and purchased a custom skin preparation kit containing hand sanitizer, two sleeves, the antiseptic, and any associated towels or sponges typically needed for the skin preparation.

After the second PDSA cycle, the committee members elected to reinforce education by training the Service Leads, defined as the nurse and scrub technician leaders for each surgical specialty. These leaders assist and guide the circulating nurses and scrub technicians in their daily activities. The frontline leadership was provided with a class on the AORN guidelines and the manufacturer's recommendations for the antiseptic use. The hope was that these leaders could reinforce best practice behaviors. One month after this class, the third and final audit commenced.

Peri-operative antibiotic optimization

We undertook a workflow analysis of peri-operative antibiotic administration. The surgeon orders the procedure through the electronic health record (EHR) and can choose to order peri-operative antibiotics at this time that are consistent with the hospital's peri-operative antibiotic guideline. However, the initial audit revealed that the antibiotic portion of the order set was not being used regularly. Instead, the attending surgeon would simply give a verbal order for the antibiotic of his or her choice just prior to incision. As a result, the antibiotic choice deviated from hospital guidelines, the dose was not checked by the pharmacy, and the administration time was suboptimal.

Denver Health's antimicrobial stewardship team provides guidance for antibiotic selection and dose for a number of common surgical procedures. These are published on the internal subsite as well as on the antimicrobial smartphone application [11]. Preferred peri-operative antibiotics are stocked in the automated dispensing machines in the operating room. These are not patient-profile specific; therefore, they are readily available to nurses without additional pharmacy review. Uncommonly used antibiotics are available to surgeons through the main inpatient pharmacy. These orders require additional pharmacist review prior to dispensing.

The PDSA process led to the development of three criteria to determine if antibiotic prophylaxis was following best practice guidelines. These criteria were defined as the antibiotic order placed at time of the surgical procedure ordering, the choice of antibiotic congruent with the hospital's peri-operative prophylaxis guidelines, and the dose of antibiotic appropriate for the patient's weight. The institution's peri-operative antibiotic prophylaxis guideline is based largely on the Surgical Infection Society's consensus document [12]. The PDSA evaluation also suggested that we needed to improve adherence to the existing EHR order sets. This set already complies with the hospital's antibiotic guidelines, but surgeons were not choosing an antibiotic when ordering the procedure. On August 28, 2017, a “hard stop” was implemented in the EHR for all surgical procedures requiring the peri-operative antibiotic to be chosen before the procedure order could be completed. A second audit was performed after the introduction of the “hard stop” to determine its impact on the three criteria.

Data acquisition

Data were obtained from multiple sources. The skin preparation audits were conducted by six nurses who were familiar with the operating room and were trained as auditors in the best practices. Data on the six required criteria were collected on a standard form. To evaluate the peri-operative antibiotic order timing, the antibiotic choice, and the antibiotic dose, data were abstracted from the EHR by manual chart review. Infection preventionists performed a combination of electronic and manual chart review to obtain SSI data. The case-finding methodology has been described in a previous publication [13]. All possible SSI were reviewed by a minimum of two infection preventionists.

Data analysis

The primary outcomes of this study were improvement in adherence to recommended skin preparation practices and to institutional guidelines for optimal antibiotic administration. The secondary outcome was SSI in the observed procedures. This quality improvement project was approved by the Quality Improvement Review Committee at Denver Health Medical Center.

Two power analyses were performed at the beginning of the project for the primary outcomes, with the threshold for adequate power set at 80% when alpha = 0.05. For the skin preparation intervention, a sample of 69 (total pre- and post-intervention) was powered adequately to detect a 20% improvement in adherence. In order to detect smaller effect sizes, more skin preparations were observed than needed for the power analysis. For the antibiotic intervention, a sample of 466 (equally split between pre- and post-intervention periods) was powered adequately to detect a 20% improvement in adherence to recommended antibiotic administration rules. All statistical analyses were performed using SAS version 9.3 (SAS Institute, Cary NC).

To determine if the SSI rate differed between the time periods, we used risk adjustment models developed by the NHSN to calculate a standardized infection ratio (SIR) for each time period. The baseline period was September 1, 2016, to December 31, 2016, while the intervention period was September 1, 2017, to December 31, 2017. The SIR uses these risk-adjustment models to predict the number of infections that would be expected during a specified time period given patient- and institution-specific factors. The number of observed infections is then compared with the number of expected infections to calculate a ratio.

Results

Skin preparation

In the first round of education, 38 of the 45 circulating staff nurses (84%) received education about skin preparation via peer-to-peer teaching, and 19 circulating nurses (42%) were present for the skin preparation in-service. Relatively few (11%; n = 5) of the circulating nurses did not receive either form of education. In the second round of education, all of the front-line nurse leadership (n = 6) participated in education about skin preparation.

Seventy-two and 52 skin preparations were observed in the baseline and intervention periods, respectively (Table 1). In the baseline period, the process most consistent with best practice included the use of gloves during skin preparation (n = 71; 99%), whereas hand hygiene prior to skin preparation and the use of sleeves were least concordant with best practice (n = 1; 1%).

Table 1.

Surgical Skin Antisepsis (“Skin Prep”) Observations

Procedure (%)	Baseline (n = 72)	Intervention (n = 52)	p^a
Hand hygiene prior to preparation,	1 (1)	25 (48)	<0.001
Sleeves worn during preparation	1 (1)	35 (67)	<0.001
Gloves worn during preparation	71 (99)	52 (100)	1.00
Correct cleansing time	34 (47)	44 (85)	<0.001
Correct drying time	48 (67)	33 (87)^b	0.02
Cleansing motion from incision to periphery	56 (78)	50 (96)	0.004

Chi-square test of association or Fisher exact test used as appropriate.

N = 38 for correct drying time during Intervention.

The intervention period showed statistically significant improvement in four of the six domains. Appropriate hand hygiene increased from 1% to 48% (p < 0.01). Sleeve use during the preparation increased from 1% to 67% (p < 0.01). Gloves being worn during the preparation remained near 100%. Compliance with the optimal cleansing time increased from 49% to 85% (p < 0.001). Allowance of the optimal drying time increased from 67% to 88% (p = 0.21). Cleansing motion that moved from the incision site to the periphery increased from 78% to 96% (p = 0.01).

Peri-operative antibiotic optimization

Peri-operative antibiotic order placement, correct antibiotic selection, and optimal antibiotic dose were evaluated in 466 surgical procedures, 224 (49%) in the baseline period and 236 (51%) in the intervention period. The median age of the patients was similar in the two time periods (pre-intervention 52 years [interquartile range {IQR} 41–61 years]; post-intervention 51 years [IQR 39–61 years]; p = 0.58), and approximately one-third of the patients were male in both groups (pre-intervention n = 76; 34%; post-intervention n = 76, 32%; p = 0.69). The case mixes in the two time periods did not differ significantly.

Improvements in peri-operative antibiotic choice, dose, and ordering were seen in almost all surgical services between the baseline and intervention periods. Overall, antibiotic prophylaxis was ordered pre-operatively in 132 cases (59%) in the baseline period and 165 (70%) in the intervention period (p = 0.02) (Table 2). Improved concordance with the prophylactic antibiotic choice also was seen after the intervention (52% versus 95%; p < 0.001). Provision of the optimal weight-based antibiotic dose improved slightly, from 91% to 94% (p = 0.22). The gynecology service had excellent adherence to the recommended peri-operative antibiotic, the appropriate dose, and use of the pre-operative order set in both the baseline and intervention periods. The use of the pre-operative antibiotic order was incrementally smallest among orthopaedic and neurosurgery providers.

Table 2.

Assessment of Peri-Operative Antibiotic Choice and Dose and Use of Pre-Operative Antibiotic Order Set in Baseline and Intervention Periods^a

Number (%)	Period	All procedures (baseline n = 224; intervention n = 236)	General surgery (baseline n = 61; intervention n = 78)	Orthopaedics (baseline n = 77; intervention n = 71)	Gynecology (baseline n = 44; intervention n = 49)	Neurosurgery (baseline n = 42; intervention n = 38)
Correct antibiotic	Baseline	116 (52)	41 (58)	8 (10)	43 (98)	24 (57)
Correct antibiotic	Intervention	225 (95)	73 (94)	70 (100)	49 (100)	32 (84)
Correct dose	Baseline	201 (91)	52 (73)	75 (97)	43 (98)	31 (74)
Correct dose	Intervention	221 (94)	69 (88)	70 (100)	48 (98)	34 (89)
Used order set	Baseline	132 (59)	37 (52)	40 (52)	43 (98)	12 (29)
Used order set	Intervention	165 (70)	70 (90)	42 (60)	46 (94)	6 (16)

Boldface type signifies a significant difference between the baseline and intervention periods.

Standardized infection ratio

In the baseline period, the hospital's combined SIR for the ten procedure types was significantly higher than the national benchmark data (SIR = 2.0; p = 0.006) (Table 3). During the intervention period, the combined SIR diminished to 1.2 (p = 0.70), which is consistent with the national benchmark data. The greatest improvement in SIR was seen for general surgical procedures.

Table 3.

Surgical Site Infection Rates in Baseline and Intervention Periods^a

	Baseline period			Intervention period
Surgery type	N	SSI	Standardized infection ratio (95% CI)	N	SSI	Standardized infection ratio
Total	475	18	2.0 (1.1–3.0)	431	11	1.2 (0.6–2.0)
General surgery (colon, breast, gastric)	83	10	2.7 (1.3–4.8)	107	4	0.8 (0.3–1.9)
Neurosurgery (craniectomy, spinal fusion)	124	3	1.6 (0.4–4.3)	84	3	2.7 (0.7–7.3)
Gynecology (vaginal or abdominal hysterectomy)	36	2	Not calculated^a	46	0	0.4 (0.0–3.0)
Orthopedic (total knee arthroplasty, total hip arthroplasty, open reduction and internal fixation of fracture)	243	4	1.4 (0.4–3.4)	209	4	1.6 (0.5–3.9)

Boldface type signifies a significant difference between the baseline and intervention periods.

CI = confidence interval; SSI = surgical site infection.

Discussion

This quality improvement project addressed adherence to best practices in skin preparation and use of an electronic “hard stop” to both standardize and optimize the administration of peri-operative antibiotic prophylaxis. We have shown statistically significant improvements in several process measures and a nonsignificant reduction in the combined SIR for the examined procedures.

Historically, skin preparation technique has been poorly studied. In fact, the AORN guidelines that require hand hygiene plus the donning of gloves and a sterile sleeve prior to the skin preparation are not substantiated with references [9]. Anecdotally, it took longer to perform an appropriate skin preparation according to the AORN and manufacturers' guidelines. Several surgeons discouraged the nurses from performing all elements, citing time constraints. This barrier to success was reduced, but not eliminated, by support from the hospital administration as well as demonstrating the positive impact on SSI rates.

We were surprised that the PDSA analyses revealed equipment to be a major barrier to appropriate skin preparation. The addition of hand sanitizers, the increase in conveniently placed surgical sleeves, and ultimately the development of a customized skin preparation kit were vital to the success of this project. These additions required capital investment by the hospital, although the estimated cost avoidance of SSI far outweighs the costs associated with facilitating adherence to best practice recommendations.

Our positive experience with the “hard stop” for antibiotic ordering correlates with the published literature. A systematic review by Davey et al. [14] found EHR-restrictive interventions such as a hard stop improved compliance with hospital antibiotic guidelines by 15%. We saw a 30% improvement in adherence to institutional guidelines for peri-operative antibiotics.

Although we were pleased with the improvement in the peri-operative antibiotic prescription order timing, choice, and dose, we were somewhat disappointed that none of the elements reached 100% adherence. Although the surgeons have one primary order set that is used to schedule surgery, there are alternative electronic pathways to order materials for an operative case without using the order set. These alternative pathways do not include the antibiotic guidelines. Additionally, surgeons are encouraged to use the recommended antibiotics detailed on the order set but are not required to. If they do not agree with the recommended dosage, they are free to change the dose in the order set. Work needs to be done with surgeons to blend their preferences further with patient safety initiatives.

There are limitations to this study. The Hawthorne effect may have biased the skin preparation results. As there was no way to hide the presence of the auditors in the evaluations, it is possible that nurses changed the performance of the skin preparation while being watched. As nurses, residents, and surgeons leave and new staff enter our hospital, we cannot be assured that these process measure improvements will be sustained. Finally, as this is a single-center quality improvement study, the results may not be applicable to other facilities that are geographically and culturally diverse.

Conclusions

Our team undertook a quality improvement project to improve skin preparation and to standardize the antibiotic prescribing process. The project resulted in the development of custom skin preparation kits and a “hard stop” to require antibiotic prescriptions at the time of surgical case orders. We are optimistic that these tangible interventions will ensure reductions in SSI in the future.

Footnotes

Author Disclosure Statement

No funding was received for this project. None of the authors has any conflict of interest to disclose.

References

Magill

, Edwards

, Bamberg

, et al. Multistate point-prevalence survey of health care-associated infections. N Engl J Med, 2014; 370:1198–1208.

The Joint Commission's Implementation Guide for NPSG. 07.05. 01 on surgical site infections: The SSI Change Project 2013. Available at: https://www.jointcommission.org/assets/1/18/Implementation_Guide_for_NPSG_SSI.pdf

Zimlichman

, Henderson

, Tamir

, et al. Health care-associated infections: A meta-analysis of costs and financial impact on the US health care system. JAMA Intern Med, 2013; 173:2039–2046.

Umscheid

, Mitchell

, Doshi

, et al. Estimating the proportion of healthcare-associated infections that are reasonably preventable and the related mortality and costs. Infect Control Hosp Epidemiol, 2011; 32:101–114.

Berrios-Torres

, Umscheid

, Bratzler

, et al. Centers for Disease Control and Prevention Guideline for the Prevention of Surgical Site Infection, 2017. JAMA Surg, 2017; 152:784–791.

Mangram

, Horan

, Pearson

, Silver

, Jarvis

. Guideline for Prevention of Surgical Site Infection, 1999. Centers for Disease Control and Prevention (CDC) Hospital Infection Control Practices Advisory Committee. Am J Infect Control, 1999; 27:97–132.

About Denver Health. Denver CO: Denver Health and Hospital Authority; 2017. Available at: http://www.denverhealth.org/about-us

Surgical Site Infection (SSI) Event. 2018. Available at: https://www.cdc.gov/nhsn/pdfs/pscmanual/9pscssicurrent.pdf

Guideline for preoperative patient skin antisepsis. Guidelines for Perioperative Practice. Denver CO: AORN; 2015. pp. 43–66.

10.

Classroom Education Program – Clinician Presentation. In: Fitzgerald L, Kurtz J, editors. CareFusion Corporation; 2014.

11.

Young

, Shihadeh

, Skinner

, et al. Implementation of an institution-specific antimicrobial stewardship smartphone application. Infect Control Hosp Epidemiol, 2018; 39:986–988.

12.

Bratzler

, Dellinger

, Olsen

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

13.

Knepper

, Young

, Jenkins

, Price

. Time-saving impact of an algorithm to identify potential surgical site infections. Infect Control Hosp Epidemiol, 2013; 34:1094–1098.

14.

Davey

, Marwick

, Scott

, et al. Interventions to improve antibiotic prescribing practices for hospital inpatients. Cochrane Database Syst Rev, 2017; 2:CD003543.