Institution Name: University of Maryland Medical Center
Submitter Name and Title: Mark Kligman, MD
Authors: Sami Tannouri, MD; Stephen Kavic, MD; Yvonne Rasko, MD; and Mark Kligman, MD
Name of the Case Study: Postoperative Pain and Nausea Protocols Decrease Length of Stay without Increasing Readmissions following Laparoscopic Gastric Bypass: A Staged Approach to a Bundled Care Pathway
Reducing the hospital length of stay (LOS) improves access to hospital beds and decreases cost. For repetitive, high-volume practices such as bariatric surgery, standardization of care can reduce resource utilization by eliminating unnecessary testing, medications, and interventions in uncomplicated cases and by early identification and treatment of complications. The resulting cost savings ultimately benefit both patients and health care systems. In states with capitated care, such as Maryland, it becomes even more important to ensure that cost reduction measures do not adversely impact patient outcomes. Standardized postoperative protocols after surgery have been implemented in bariatric surgery to decrease LOS. In bariatric surgery, postoperative analgesic and antiemetic protocols have not yet been standardized.1-4
At the University of Maryland, prior to instituting analgesic or nausea protocols, about two-thirds of patients undergoing laparoscopic bypass surgery were discharged on postoperative day 1 (POD 1). Most remaining patients stayed additional days due to inadequate oral intake due to persistent nausea. Moreover, the use of narcotic analgesics appeared to be a significant exacerbating factor for nausea. To address this issue, we developed a narcotic-sparing analgesic protocol and an antiemetic protocol. These protocols have preoperative, intraoperative, postoperative, and post-discharge components. In addition, because these protocols have multiple independent interventions, they are bundled interventions. Typically, all interventions in patient care bundles are implemented simultaneously, preventing analysis of the efficacy of individual interventions. The downside of this approach is that ineffective (and often expensive) interventions are not identified for potential elimination.5-7 We sought to implement a standardized postoperative analgesia and nausea bundle in a sequential stages, allowing us to analyze each intervention as it was added. We hypothesized that our protocols would lead to improved pain and nausea control, which would allow decreased hospital LOS.
Prior to implementation of our protocols, all patients followed a standardized care pathway. This pathway included routine use of patient-controlled anesthesia, discharge on oral narcotics, and an inpatient order set for nausea, which was written with minimal guidance or education for the administering nurses.
From 2015 to 2018, components of the protocols were sequentially introduced in stages. This required buy-in from the surgeons, as well as engagement of the nurses led by nursing leadership both in the preoperative and postoperative settings. Intraoperative protocol changes required coordination with anesthesia providers as well as our surgical technologists.
Development and selection of interventions came from a combination of medical literature as well as provider consensus. Discussion of practice patterns across many bariatric surgeons across the country yielded us a list of interventions, which our group discussed, modified, and selectively implemented. Enacting these protocols was supported and encouraged by divisional and departmental leadership, and changes were tracked both via internal quality data and coded metrics as well as MBSAQIP data.
From 2015 to 2018 key steps were sequentially introduced in stages. Throughout implementation, in the outpatient preoperative setting patients were introduced to and taught an educational module with extensive details of their upcoming care. This included education on their inpatient medication regimen for pain and nausea such that they would be able to appropriately ask for treatment for their symptoms in a timely and effective manner after their surgeries.
In our first implemented stage, on the day of surgery prior to the operation, each patient is given a scopolamine patch. Intraoperatively, anesthesia administered 1g IV acetaminophen. Our first postoperative intervention was to eliminate patient-controlled analgesia (PCA), which required engagement of anesthesia and the pain service as well as our floor nurses. Nursing was engaged via in-service teaching facilitated by charge nurses as well as the surgeons. The next step was implementation of a global reduction in postoperative narcotic utilization, again by re-education of the nurses on the bariatric patient unit and outpatient expectation-setting with our patients. Along with the intraoperative acetaminophen, patients are given scheduled PO acetaminophen, PRN oxycodone, and rescue PRN morphine based on strict pain score guidelines.
Patients were discharged with 10 tabs of 5mg oxycodone.
Our second stage introduced intraoperative transversus abdominal plane (TAP) and rectus sheath blocks using 60 mL 0.25 percent Marcaine with 1 percent epinephrine expanded to 90 mL administered by the surgeon intraoperatively via laparoscopic visualization.8-9 Our patients were discharged with only 5 tabs of 5mg PO oxycodone.
During our third stage we introduced an aggressive standardized anti-nausea protocol, which required engagement of both preoperative and postoperative nursing staff on the proper sequence and timing of ondansatron, haldol, and diphenhydramine. This required implementing a nausea score that would guide the administration of antiemetic. The house staff was briefed on this protocol as to not disrupt the protocol with adjunct anti-emetics. Finally, we held intensive bariatric unit nursing education and expectation-setting on post-operative ambulation and diet, implementing an inpatient checklist with hourly walking and oral intake goals. Additionally, each sequential step required physician-led education of the house officers, as well as changing electronic order sets via help from bariatric office coordinators and the university’s information technology (IT) department.
Staff
These interventions were primarily driven by two surgeon champions and involved the entire staff of our bariatric nursing unit, which was educated by our nursing champion. Office staff within the bariatrics program, particularly our bariatric program manager who helped implement the IT changes as well as the division’s clinical reviewer, were integral.
Costs
Minor additional costs were incurred intraoperatively to administer bupivacaine transversus abdominus plane blocks. MBSAQIP membership was a fixed cost already within the division’s budget.
Budget
There was no additional funding source for this intervention.
We retrospectively analyzed data collected via MBSAQIP on patients undergoing laparoscopic roux-en-y gastric bypass from 2014 to 2018. These patients were broken up into groups based on each stage’s implementation dates. Length of stay data was analyzed via t-test and readmissions were analyzed by Fischer exact test. Our length of stay decreased significantly with implementation of the full bundle compared to baseline (1.66 days vs. 1.33 days, p = 0.027). Each individual intervention was then analyzed against baseline as well as the previous intervention. When each stage was individually evaluated, introduction of TAP blocks had a significant impact on LOS, decreasing from 1.79 to 1.5 days (p=0.04). Our 30-day readmission rate throughout the entire implementation decreased slightly, though not significantly (5.6% vs. 3.6%, p=0.14). We were thus able to reduce length of stay without increasing readmissions. Unfortunately, we were unable to capture consistent pain or nausea scores largely due to the challenges of implementation of a new electronic medical record. Even though we had no direct measure of pain or nausea during our interventions in this austere data environment, we knew 75 percent of readmissions were pain and/or nausea related. Thus, particularly with early readmissions, our protocol likely was effective in treating our postoperative patients’ pain and nausea. Patients were able to go home earlier and returned to the hospital at the same rate as we were better able to control their pain and nausea postoperatively.