2019, University of Utah Health was identified (American College of Surgeons’ National Surgical Quality Improvement Program, NSQIP) as an outlier with an unacceptably high rate of surgical site infections (SSIs) following a colectomy. During a colectomy, we remove a part or all of the patient’s large intestine. As SSIs are the leading cause of readmission to the hospital following a surgery, and are a significant cause of mortality after surgery, we felt that reducing this rate was important to ensuring that patients receive high-quality surgical care at the University of Utah. 

Defining the problem 

We formed an interdisciplinary team led by Luke Martin, one of the colorectal surgeons, and Hailey Harris, an infection preventionist. We also included OR teams, nursing, anesthesia, and pharmacy. Following the identification of this problem, our initial approach was two-fold.  

First, we performed a comprehensive review of the literature to identify best practices for reducing SSIs in post-colectomy patients. We reviewed all recent post-colectomy SSIs in our institution, and we paid special attention to the operative reports and whether best practices already established in the literature were followed.  

We then used quality improvement methodology to perform a multi-factorial analysis of possible contributing factors to post-colectomy SSIs across multiple domains. The below fishbone diagram was most helpful in guiding next steps.  

We explored factors included such as:  

1. Patient differences, like elevated A1C (a test measuring average blood sugar levels) and comorbidities

2. Operative methods factors, such as not using a wound protector during surgery, issues with bowel prep and not changing instruments and gloves after a contaminated portion of the case

3. Provider factors, like the misdiagnosis of SSIs 

4. Environment or technology  

5. Cultural factors, such as feeling safe to speak up 

The change 

We approached each of the areas of opportunity on our fishbone diagram. 

1. Optimizing preoperative care  

We began scheduling all patients who were having a colectomy in a pre-surgical optimization clinic. Here, we collect an HgA1c if the patient hasn’t had one within 3 months. We use a patients’ HgA1c as part of our decision-making on when to operate on patients without urgent surgical indications such as cancer, as poor glucose control after surgery has been shown to contribute to SSIs. 

We also implemented the pre-operative administration of an Ensure Pre-Surgery Clear Carbohydrate drink to promote normoglycemia (normal blood sugar levels). The drink is given to all patients in the pre-op area two hours before surgery. Collaboration was required between anesthesia, pre-op nursing, and pharmacy to modify our NPO (no eating or drinking) guidelines prior to surgery, and to make sure this product was available and properly administered.  

2. Addressing diagnosis patterns for patients at home  

An SSI is the leading cause of readmission to the hospital, so providers taking care of recently discharged patients are particularly vigilant as they look for signs of infection. We met with the Huntsman at Home providers, who care for cancer patients after surgery when they get home. We found that out of an abundance of caution, that nonsurgical providers were overprescribing antibiotics due to wound concerns, qualifying patients as having developed an SSI when this was not clinically accurate. We developed a protocol for home providers to send photos of the incision to the operating surgeon for collaborative review prior to prescribing antibiotics for wound concerns unless the patient is clinically ill. We implemented a similar protocol for residents.  

3. Standardizing operative methods across provider groups  

A major challenge of this project is that colectomies are performed by many surgeons at this institution. Providers from colorectal surgery, trauma and emergency general surgery, surgical oncology, and occasionally gynecologic oncology perform colectomies.   

We worked with multiple surgical teams to standardize our institutional approach to colectomies. Our standard approach includes the use of a wound protector for all colectomies and a standardized bowel prep protocol. These steps were added as an order set in Epic, to minimize variation between providers. We also added the use of Zofran, an anti-nausea medication, so the pre-operative bowel prep protocol was better tolerated by patients. We standardized the use of a clean closure following contaminated parts of a procedure. We also provided more guidance about performing off-midline incisions, which have been shown to have a lower rate of surgical site infections.  

The results of these changes made prevention of SSIs like a checklist for our care teams, which in turn has reinforced best practices.   

4. Examining environmental factors  

We examined factors such as adjusting the room temperature inappropriately so that patient normothermia (body temperature) wasn't maintained. 

Before this project, a surgeon, anesthesiologist or OR nurse could turn down the temperature if they felt hot. But that can affect patient normothermia (body temperature), leading to increased risk of infection. Now, when someone says they feel warm, it initiates a multidisciplinary discussion and makes the providers talk about whether adjusting the room temperature is the right move for the patient.  

Results  

While the number of colectomies being performed is increasing (the blue line at the top of the graph), the linear rate of total surgical site infections post-colectomy is falling. This project has reduced superficial surgical site infections as well.  

Sustaining the Results and Future Directions 

Keeping up with other interventions over the course of years has been challenging. Many services offer colectomies, and they are performed at several ORs throughout the system. This raises issues with equipment availability, different OR cultures, different provider preferences, and keeping up with provider education. Once an intervention has been implemented, people tend to backslide over time. We are figuring out how to mitigate this effect. We’ve been focused on refining our data tracking and building in processes within the EMR (electronic medical record) so that we can share more with our teams on whether or not the interventions continue to work. It is a learning process that we continue to refine.