In Pursuit of an Opioid-Free Pediatric Ambulatory Surgery Center: A Quality Improvement Initiative : Anesthesia & Analgesia

Secondary Logo

Journal Logo

Original Research Articles: Original Clinical Research Report

In Pursuit of an Opioid-Free Pediatric Ambulatory Surgery Center: A Quality Improvement Initiative

Franz, Amber M. MD, MEng; Martin, Lynn D. MD, MBA; Liston, David E. MD, MPH; Latham, Gregory J. MD; Richards, Michael J. BM; Low, Daniel K. BM, BS

Author Information
doi: 10.1213/ANE.0000000000004774
  • Free

Abstract

KEY POINTS

  • Question: Is it possible to minimize perioperative opioids in pediatric ambulatory surgical patients without compromising patient outcomes or value?
  • Findings: During an 18-month-long quality improvement project, our team reduced intraoperative opioid administration from 84% to 8% and postoperative morphine administration from 11% to 6% using a combination of dexmedetomidine, nonsteroidal anti-inflammatory drugs, and regional anesthesia.
  • Meaning: Through a combination of accessible perioperative outcome data, regular data review, and a culture that embraces continuous improvement, our team was able to minimize perioperative opioids in our pediatric ambulatory surgical patients without compromising patient outcomes or value.

INTRODUCTION

Problem Description

Seattle Children’s Bellevue Clinic and Surgery Center is a standalone pediatric clinic and ambulatory surgery facility that cares for >4000 surgical patients annually.1 The most commonly performed surgeries account for >60% of the annual case volume (Table 1). Anesthesiologists and nurse anesthetists follow standardized anesthesia protocols to optimize consistency of care for these high-volume surgeries.

Table 1. - Standardized Intraoperative Anesthesia Protocols for Bellevue Clinic and Surgery Center’s Most Common Surgeries as of December 2017 and June 2019
Procedure Opioid-Inclusive Protocols (2017) Opioid-Free Protocols (2019)
Otolaryngology Myringotomy with tympanostomy tubes Fentanyl 1 µg/kg intranasal (intraoperatively) Ibuprofen 10 mg/kg oral (preoperatively)
Tonsillectomy and adenoidectomy/tonsillectomy Morphine 0.1 mg/kg Dexmedetomidine 1 µg/kg
Acetaminophen 15 mg/kg Ketorolac 0.5 mg/kg (max 30 mg)
Dexamethasone 0.15 mg/kg (max 4 mg) Dexamethasone 0.5 mg/kg (max 8 mg)
Adenoidectomy Morphine 0.05 mg/kg Dexmedetomidine 0.5 µg/kg
Acetaminophen 15 mg/kg Ketorolac 0.5 mg/kg (max 30 mg)
Dexamethasone 0.15 mg/kg (max 4 mg) Dexamethasone 0.5 mg/kg (max 8 mg)
Urology Circumcision/buried penis repair/hypospadias repair <3 y—single shot caudal <3 y—pudendal or single shot caudal
>3 y—penile block by surgeon >3 y—pudendal block
Fentanyl 0.5–1 µg/kg Pudendals <15 kg
 Ropivacaine 0.2% (max 5 mL/side)
Pudendals >15 kg
 Ropivacaine 0.5% (max 5 mL/side)
Dexmedetomidine 0.5–1 µg/kg
Orchiopexy/inguinal hernia repair/hydrocelectomy <3 y—single shot caudal <3 y—single shot caudal
>3 y—ilioinguinal block
>3 y—ilioinguinal block  Ropivacaine 0.5% 0.1–0.2 mL/kg
 Ropivacaine 0.5% 0.1–0.2 mL/kg Dexmedetomidine 0.5–1 µg/kg
Fentanyl 1–2 µg/kg Ketorolac 0.5 mg/kg (max 30 mg)
Ketorolac 0.5 mg/kg (max 30 mg)
General surgery Inguinal hernia repair/hydrocelectomy <3 y—single shot caudal <3 y—single shot caudal
>3 y—ilioinguinal block >3 y—ilioinguinal block
 Ropivacaine 0.5% 0.1–0.2 mL/kg  Ropivacaine 0.5% 0.1–0.2 mL/kg
Fentanyl 1–2 µg/kg Dexmedetomidine 0.5–1 µg/kg
Ketorolac 0.5 mg/kg (max 30 mg) Ketorolac 0.5 mg/kg (max 30 mg)
Umbilical hernia repair/epigastric hernia repair Rectus sheath block Rectus sheath block
 Ropivacaine 0.5% 0.2 mL/kg/side  Ropivacaine 0.5% 0.2 mL/kg/side
Fentanyl 0.5–1 µg/kg Dexmedetomidine 0.5–1 µg/kg
Ketorolac 0.5 mg/kg (max 30 mg) Ketorolac 0.5 mg/kg (max 30 mg)
Orthopedics Knee arthroscopy and meniscus repair Adductor canal block Adductor canal block
 Ropivacaine 0.5% 0.1–0.2 mL/kg  Ropivacaine 0.5% 0.1–0.2 mL/kg
Fentanyl 0.5–2 µg/kg Dexmedetomidine 1–2 µg/kg
Ketorolac 0.5 mg/kg (max 30 mg) Ketorolac 0.5 mg/kg (max 30 mg)
Knee arthroscopy and anterior cruciate ligament repair Adductor canal catheter Adductor canal catheter
 Ropivacaine 0.5% 0.1–0.2 mL/kg bolus + 0.2% infusion × 3 d  Ropivacaine 0.5% 0.1–0.2 mL/kg bolus + 0.2% infusion × 3 d
Sciatic nerve block Sciatic nerve block
 Ropivacaine 0.5% 0.2 mL/kg  Ropivacaine 0.5% 0.2 mL/kg
Fentanyl 0.5–2 µg/kg Dexmedetomidine 1 µg/kg bolus + 0.5 µg/kg/h infusion
Ketorolac 0.5 mg/kg (max 30 mg) Ketorolac 0.5 mg/kg (max 30 mg)
Trigger thumb repair Median nerve block by surgeon Median nerve block by surgeon
Fentanyl 0.5–1 µg/kg Dexmedetomidine 0.5–1 µg/kg
Dermatology Pulse dye laser Fentanyl 0.5–1 µg/kg Dexmedetomidine 0.5–1 µg/kg
Ketorolac 0.5 mg/kg (max 30 mg) Ketorolac 0.5 mg/kg (max 30 mg)
Ophthalmology Strabismus repair Fentanyl 1–2 µg/kg Dexmedetomidine 1–2 µg/kg
Ketorolac 0.5 mg/kg (max 30 mg) Ketorolac 0.5 mg/kg (max 30 mg)
Dexmedetomidine, fentanyl, ketorolac, dexamethasone, and acetaminophen were administered intravenously intraoperatively if not otherwise specified. Ropivacaine was administered neuraxially or perineurally after induction and before surgery, with a maximum dose of 3 mg/kg if not otherwise specified.
Abbreviation: Max, maximum.

Since opening in 2010, Bellevue has utilized multimodal analgesia to minimize pain and perioperative opioid use, including regional anesthesia for a large number of protocols (Table 1). Despite these efforts, opioid use increased in 2017 (Figure 1), with 84% of surgical patients receiving intraoperative opioids in December 2017. In early 2018, 2 problems converged that led our team to prioritize opioid-related quality improvement work:

F1
Figure 1.:
Number of opioid ampules used intraoperatively at Bellevue Clinic and Surgery Center each month between January 2017 and June 2019.
  1. A national shortage of intravenous opioids pressured health care systems to alter protocols to conserve supplies.2–4
  2. The national opioid crisis deepened, as evidenced by opioid overdose fatalities (47,600 deaths) exceeding deaths caused by motor vehicle accidents.5,6 Opioid use following routine surgery became a recognized gateway to opioid misuse disorder.7–10

Available Knowledge

Most surgical patients in the United States receive opioids perioperatively.11 Opioids provide effective analgesia with a rapid onset of action and are considered a component of a balanced anesthetic technique.12 The acute side effects of opioids include respiratory depression, pruritus, urinary retention, nausea, vomiting, constipation, and altered mental status.9,11

In addition to these side effects, intraoperative opioid administration has been variably associated with increased postoperative opioid requirements. Chia et al13 described higher postoperative pain scores and analgesic requirements in total abdominal hysterectomy patients who received high- versus low-dose fentanyl. Fentanyl has also been found to increase hyperalgesia caused by a subcutaneous electric burn.14 Similarly, remifentanil can cause opioid-induced hyperalgesia.15,16 However, a systematic review of intraoperative opioid-induced hyperalgesia found that there were insufficient data to draw a conclusion about fentanyl and morphine.16

The long-term effects of perioperative opioids are also concerning. A population study of 36,177 adult elective surgical patients included 80% undergoing minor surgery and 20% undergoing major surgery.7 The rate of new persistent opioid use, defined as opioid prescription fulfillment 90–180 days following surgery, was similar between groups (5.9%–6.5%). Claims data from 88,637 patients aged 13–21 years undergoing a range of surgical procedures showed a 4.8% (range, 2.7%–15.0%) incidence of new persistent opioid use versus 0.1% in a nonsurgical control group.8

Rationale

It eventually became apparent to anesthesiologists and nurse anesthetists at Bellevue that perioperative opioids should be minimized. However, this realization was not immediate but emerged over time.

Our anesthesia team is focused on providing value to patients, which involves improving or maintaining outcomes while reducing costs. We meet regularly to make changes to our standardized protocols with this in mind. After examining patient outcome data from 2017, we discovered that intravenous acetaminophen was not helpful in certain surgeries and so opted to replace this expensive medication with a cheaper alternative. After reviewing the literature, we identified dexmedetomidine as a possible replacement.

Dexmedetomidine is a highly selective α-2 agonist that reduces volatile anesthetic requirements, provides anxiolysis, reduces emergence delirium, and provides analgesia with minimal risk of respiratory depression.17 Postoperatively, dexmedetomidine reduces postoperative pain scores and analgesic requirements for patients undergoing a range of ambulatory surgeries.18,19 Side effects include bradycardia and hypotension.17

Around the time that we were evaluating dexmedetomidine as a replacement for intravenous acetaminophen, the national opioid shortage developed.2–4 In early 2018, a hospital-wide e-mail alerted staff of Seattle Children’s critically low supply of intravenous opioids, with a request to conserve supplies. This impetus prompted the Bellevue team to develop a new standardized protocol that eliminated intraoperative opioids for one of our highest-volume surgeries, tonsillectomy and adenoidectomy. After much discussion, we replaced intraoperative morphine and acetaminophen with dexmedetomidine and ibuprofen.20

Our success with reducing perioperative opioid administration for tonsillectomy and adenoidectomy without compromising effective analgesia inspired us to expand the use of dexmedetomidine at our facility. As the long-term effects of perioperative opioids became known,5,7–10 we began to incorporate dexmedetomidine into other surgeries, while exploiting opioid-sparing analgesics already popular at Bellevue, such as regional anesthesia and nonsteroidal anti-inflammatory drugs. The results of these interventions were so promising that, by January 2019, we decided to remove opioids from all standardized intraoperative anesthesia protocols.

Specific Aims

We sought to develop effective anesthesia protocols that minimize perioperative opioids for ambulatory pediatric surgeries and add value to our services by maintaining or improving perioperative outcomes while reducing costs.

METHODS

Context

Bellevue Clinic and Surgery Center is a satellite campus of Seattle Children’s located in Bellevue, Washington, where routine, noncomplex ambulatory surgeries are performed on relatively healthy patients. Bellevue functions as a Learning Health System and was built on lean principles, including unity of purpose, continuous improvement, visual tracking of information, standardization of best practices, value creation and waste minimization, and respect for workers.21,22 Bellevue’s leaders and frontline staff are engaged in and supportive of continuous improvement and standardized care processes—team members meet regularly to review current protocols and update them with evidenced-based practice. Once a new protocol is deployed, compliance is encouraged by using customized, case-specific checklists built into the anesthesia module of the electronic medical record. Data from the electronic medical record are stored in the hospital’s electronic data warehouse. Clinical teams have immediate access to perioperative data and analytics through MDmetrix OR Advisor (Seattle, WA), a software program that accesses aggregate deidentified data from the warehouse. Clinical teams can query continuously updated data, create custom patient cohorts, and surface key metrics in the form of statistical process control charts, allowing for near real-time understanding of how processes are performing and the ability to detect clinical improvement as a result of protocol changes. Thus, rapid Plan-Do-Study-Act cycles23 and effective quality improvement efforts are possible due to accessible data, regular data review, and timely communication among team members via e-mails, bulletins, and meetings.

Interventions

In 2018, our anesthesia team began iterative steps to reduce perioperative opioid use. By January 2019, we removed opioids from all protocols, based in part on the success of our opioid-sparing tonsillectomy and adenoidectomy protocol as described by Franz et al.20 Changes to standardized protocols for procedures with the greatest monthly surgical volume were prioritized. Table 1 summarizes the protocols in place in December 2017 and June 2019; however, it does not show the Plan-Do-Study-Act cycles that occurred in the interim. The major changes across all protocols were to substitute intraoperative fentanyl and/or morphine with dexmedetomidine and replace intravenous acetaminophen with ketorolac. The tonsillectomy and adenoidectomy protocol had the most Plan-Do-Study-Act cycles (4) between December 2017 and June 2019.

At Bellevue, most surgical patients receive a mask induction with sevoflurane, propofol bolus, lactated Ringer solution, dexamethasone, and ondansetron. Children undergoing myringotomy and tympanostomy tubes are the exception, as they undergo surgery without intravenous access. Premedication with midazolam is rare (<1%); our induction rooms allow for parental presence during induction, and a child life specialist is available if needed.

Measures

We tracked the number of opioid ampules used intraoperatively at Bellevue Clinic and Surgery Center each month to identify opioid usage patterns. We also compared the percentage of surgical patients receiving intraoperative opioids and postoperative morphine at the end of our preintervention (December 2017) and postintervention (June 2019) periods to identify changes.

For our statistical process control chart analyses, morphine rescue rates and maximum pain scores in recovery were selected as primary outcome measures to assess the efficacy of analgesic protocols. Morphine is the primary rescue medication used in our postanesthesia care unit, along with oral acetaminophen. Oxycodone is infrequently used. Pain scores were recorded by recovery nurses using either the Faces, Legs, Activity, Cry, Consolability tool, Faces Pain Scale–Revised, or a numerical Visual Analog Scale depending on patient age and developmental maturity.24,25 Recommended ages for each pain assessment tool are 0–3 years, 3–6 years, and 7+ years, respectively. Each score was converted to an 11-point (0–10) scale for the maximum pain score chart.

The balancing measures of total minutes in recovery and total anesthesia minutes were selected because dexmedetomidine has been shown to delay wake-up and discharge, which are undesirable in an ambulatory surgery setting.26 Standardized discharge criteria based on the modified Aldrete scoring system (or return to patient baseline) are used.27 Patients’ surgical sites must be stable before discharge. Total anesthesia minutes were defined as time from anesthesia start to end after handoff of an extubated patient.

Postoperative nausea and vomiting rescue rate was monitored to assess for change after removing opioids from our protocols. Medication was administered according to nurse judgment, with anesthesiologists consulted as appropriate.

Intraoperative analgesic costs were calculated to determine whether protocol changes affected costs. Local anesthetics were not included, because use of regional anesthesia remained stable from 2017 to 2019.

Analyses

Between January 2017 and June 2019, we tracked the number of opioid ampules used intraoperatively at Bellevue Clinic and Surgery Center each month using data from pharmacy.

The percentage of patients receiving intraoperative opioids in December 2017 versus June 2019 were calculated by dividing the number of intraoperative opioid ampules by the number of surgical patients for each of these months. The percentage of patients receiving postoperative morphine in December 2017 versus June 2019 were obtained from the appropriate statistical process control chart.

Data for the statistical process control chart analyses were extracted from Seattle Children’s data warehouse for all patients undergoing surgery at Bellevue from January 2017 to June 2019 using MDmetrix OR Advisor. Before pulling data from the warehouse into MDmetrix, a validation step was performed on a random sample of patients to verify data accuracy. Incomplete data were not imported.

Statistical process control charts were used to visualize continuously updated data and support rapid changes for improvement. These charts monitor a process over time and help distinguish changes due to special circumstances (ie, special cause variation, such as improvement from a protocol change) from random variation inherent in a system.28 Charts were interpreted using Shewhart's theory of variation—X-bar charts display measures consisting of continuous data with multiple values per subgroup, and P charts display dichotomous data expressed as a percentage or proportion of the total data in a subgroup.28,29 X-bar charts were used to display maximum pain score in recovery, total postanesthesia care unit minutes, and total anesthesia minutes. Each point on our X-bar charts represents the mean value for the selected measure for all surgical patients at our facility for a given month. The weighted average of these monthly means is plotted as the center line on the figures. P charts were used to display the percentage of patients who required rescue morphine and nausea medication in recovery. Each point on the P charts represents the mean percentage of all surgical patients at our facility who achieved the selected measure for a given month. The weighted average of these monthly mean percentages is plotted as the center line on the figures.

Upper and lower control limits were set at 3 sigma; 99.7% of data within a normally distributed data set are expected to fall within this range. For the X-bar charts, the sigma value was determined by averaging the standard deviations for each of the monthly subgroups. For the P charts, sigma was determined by calculating the theoretical standard error of the binomial distribution. Twelve subgroups were considered acceptable to establish a stable state.28 The presence of special cause variation suggesting a change or improvement in the system was identified using standard rules, including a single point located outside the control limits, 8 consecutive points above or below the mean, a trend of 6 consecutive increasing or decreasing points, and 2 of 3 consecutive points located near the outer one-third of the control limit.28

Intraoperative analgesic costs were calculated using September 2019 drug acquisition prices from pharmacy, case volumes, and administration data from MDmetrix for intravenous acetaminophen, morphine, fentanyl, alfentanil, dexmedetomidine, and ketorolac.

Ethical Considerations

This quality improvement project was submitted to Seattle Children’s internal review board and was not deemed a research study; thus, no further review or approval was recommended.

RESULTS

Results

Intraoperative opioid ampule use at Bellevue Clinic and Surgery Center between January 2017 and June 2019 is shown in Figure 1. Between December 2017 and June 2019, our team reduced intraoperative opioid administration from 84% (297 ampules among 352 surgical cases) to 8% (32 opioid ampules among 393 cases) and postoperative morphine administration from 11% to 6%.

Between January 2017 and June 2019, 10,948 surgeries were performed at Bellevue, with 215 (2.0%) excluded due to missing data (Table 2). Of the 10,733 cases included in the statistical process control chart analyses, 3934 occurred in 2017, 4390 occurred in 2018, and 2409 took place between January and June of 2019. Key protocol changes are annotated on the charts. Because otolaryngology surgeries comprise 40% of total case volume, changes to these protocols had a large effect on our measures.

Table 2. - Surgical Patient Demographics
January 2017 to December 2017 (Case Number = 3934) January 2018 to June 2019 (Case Number = 6799)
Sex
 Male 2438 (62.0%) 4254 (62.6%)
 Female 1496 (38.0%) 2545 (37.4%)
Mean age (y) 6.2 (0–28) 6.0 (0–23)
Mean weight (kg) 27.7 (5.7–131.3) 27.3 (4.4–132.4)
Mean body mass index (kg/m2) 18.1 (12.0–47.0) 18.1 (9.2–47.3)
American Society of Anesthesiologists physical status
 I 1974 (50.2%) 3496 (51.4%)
 II 1850 (47.0%) 3120 (45.9%)
 III 104 (2.6%) 178 (2.6%)
 Other 6 (0.2%) 5 (0.1%)
Patient ethnicity
 Non-Hispanic 3088 (78.5%) 5338 (78.5%)
 Hispanic 629 (16.0%) 1092 (16.1%)
 Patient refused 206 (5.2%) 355 (5.2%)
 Unknown 11 (0.3%) 14 (0.2%)
Patient race
 Caucasian 2283 (58.0%) 3854 (56.7%)
 Asian 352 (9.0%) 568 (8.4%)
 2 or more races 273 (6.9%) 511 (7.5%)
 Patient refused 252 (6.4%) 442 (6.5%)
 African American 192 (4.9%) 336 (4.9%)
 Other 582 (14.8%) 1088 (16.0%)
Values expressed as numbers and percentages or mean and range.

Figure 2A shows the X-bar chart for monthly mean maximum pain scores in recovery. For 2017 patients, the mean is 2.5. For the 2018–2019 cohort, the mean is 2.7. Both groups have stable processes with no special cause variation signals, indicating that any monthly variation is random or inherent in the system. There is no signal between cohorts. Note the last 6 points between January and June 2019—maximum pain scores remain stable despite removal of opioids from all intraoperative protocols.

F2
Figure 2.:
Primary outcome measures. A, Maximum pain score in the postanesthesia care unit—X-bar chart. B, Percent of patients requiring rescue morphine in recovery—P Chart. LCL indicates lower control limit; UCL, upper control limit.

The P chart in Figure 2B shows the percentage of patients who required rescue morphine in recovery. The 2017 cohort has a mean of 8.7%; there is a single signal highlighted, indicating that the process is unstable. The 2018–2019 cohort has a mean of 9.0% and is also unstable, with 3 breaches of the upper control limit; we suspect that the signals in June and August 2018 are related to recovery nurse apprehension about omitting intraoperative morphine from the tonsillectomy and adenoidectomy protocol. As the team adjusted to and became more comfortable with the use of dexmedetomidine, morphine rescue rates declined.20 By September 2018, the morphine rescue rate starts a sustained downward shift, which persists after January 2019, when intraoperative opioids were removed from all protocols. However, lack of stability prevents us from drawing conclusions when comparing cohorts.

The X-bar chart for monthly mean total postanesthesia care unit minutes (Figure 3A) demonstrates stability throughout 2017, with a mean of 62.4 minutes. Excluding June 2019, the 2018–2019 cohort also shows stability with a mean of 63.5 minutes. Special cause variation in June 2019 is explained by a high number of complex cases that were diverted to Bellevue due to the closing of multiple Seattle Children’s main campus operating rooms. These complex patients required longer recovery times than the typical Bellevue population. We displayed this June point on the chart, but its value does not contribute to the calculation of the control limits or mean given the atypical circumstances.

F3
Figure 3.:
Balancing and other measures. A, Total postanesthesia care unit minutes—X-bar chart. B, Total anesthesia minutes—X-bar chart. C, Percentage of patients requiring rescue nausea and vomiting medication in recovery—P chart. LCL indicates lower control limit; UCL, upper control limit.

The X-bar chart for monthly mean total anesthesia minutes (Figure 3B) demonstrates process stability, with means of 60.9 minutes for 2017 and 58.9 minutes for 2018–2019. There are no special cause variation signals within or between cohorts. The June 2019 point is treated similarly as above.

Figure 3C shows that the mean percentages of 2017 (1.4%) and 2018–2019 (0.9%) patients requiring rescue medication for postoperative nausea and vomiting are stable and low. Soon after implementing the opioid-sparing tonsillectomy and adenoidectomy protocol, there is a sustained reduction compared to the 2017 mean, indicating improvement.

Annual intraoperative analgesic costs decreased from approximately $85,000 in 2017 to $6000 for the first half of 2019 ($12,000 annually) using September 2019 drug prices, largely due to reduced intravenous acetaminophen administration ($19 per patient savings).

DISCUSSION

Summary

Between December 2017 and June 2019, our team reduced intraoperative opioid administration from 84% to 8% and postoperative morphine administration from 11% to 6%. Eighteen months after beginning this quality improvement project, postoperative nausea and vomiting rescue rates improved, while maximum pain scores in recovery, total anesthesia minutes, and postanesthesia care unit minutes remained stable per control chart analyses. Conclusions could not be drawn for postoperative morphine rescue rates due to process instability for this measure. Annual intraoperative analgesic costs declined by $73,000 between 2019 and 2017.

Our quality improvement project has several strengths. First, we were able to perform rapid Plan-Do-Study-Act cycles due to accessible, continuously updated data, allowing for near real-time understanding of process performance and the ability to detect improvements from protocol changes. Second, we expanded our protocols to other surgeries since publishing on our prior quality improvement project, which focused only on patients undergoing tonsillectomy and adenoidectomy surgeries.20 Third, we identified cost savings for our intraoperative analgesic medications. Finally, we identified a number of effective opioid-sparing protocols for the most common pediatric ambulatory surgeries performed at our facility among a large cohort of patients.

Interpretation

Our experience is consistent with the literature. The benefits of opioid-free anesthesia and dexmedetomidine as an analgesic adjunct and opioid-sparing agent are well described.18,19,30 The morphine rescue rate in our recovery unit declined as intraoperative opioids were replaced with dexmedetomidine, although lack of process stability prevented us from drawing conclusions. Our finding that dexmedetomidine reduces postoperative nausea and vomiting is reflected in the literature.30,31 Likewise, the time our patients spend in the postanesthesia care unit is similar to or less than prior reports.32,33

Obstacles to improve work included team member participation in standardized protocols and recovery nurse apprehension about eliminating intraoperative opioids. These challenges were overcome using technology, data, communication, education, time, and a culture that emphasizes willingness to change. Technology permitted protocols to be embedded into the anesthesia record as customizable checklists. MDmetrix allowed the team to visualize data in the form of statistical process control charts and track performance over time. Presentations of charts at weekly meetings helped educate and engage team members in thoughtful discussions around improvement. The creation of a workplace culture that supports continuous improvement started from the top down, with recruitment of staff who are open to change. Over time, the team became used to frequent evidence-based protocol updates, and change became the norm.

Limitations

This quality improvement initiative has several limitations. First, despite removing opioids from all standardized protocols, there are several surgeries that are not protocolized, and our intraoperative opioid administration rate has not yet reached zero.

Second, we selected statistical process control charts to evaluate our measures because this analytical tool provides a clear visual display of changes over time, allowing for prompt decision-making locally.29,34 However, we did not perform any supplemental statistical analyses, such as segmented regression of an interrupted time series, to adjust for confounders, generate effect estimates, and allow for determination of impact assessment.34,35 This is because (1) we are not attempting to make inferences about populations outside our facility and (b) data access for such analyses falls outside the purview of our internal review board determination.

Third, we did not directly examine hemodynamic data. However, ephedrine and glycopyrrolate administration can be used as proxies for bradycardia and/or hypotension, the most common side effects of dexmedetomidine—in 2017, 2018, and the first half of 2019, 2.7%, 2.7%, and 3.6% of patients received ephedrine, while 3.1%, 2.7%, and 1.5% received glycopyrrolate, respectively.

Fourth, we converted 3 different pain assessment tools into an 11-point (0–10) scale, which is not ideal because pain may be assessed differently depending on who evaluates pain.36 However, when separating patients by age and thus most likely pain assessment tool used, we found no special cause variation between age groups preintervention (2017) versus postintervention (2019).

Finally, and most importantly, we have minimal data on postdischarge pain and opioid use for the first week following surgery. Thus, it is not known whether there was a change in home opioid administration once intraoperative opioids were replaced with dexmedetomidine—this represents the next phase of our quality improvement initiative.

CONCLUSIONS

This quality improvement project sought to minimize perioperative opioids for all pediatric ambulatory surgeries. By combining dexmedetomidine with nonsteroidal anti-inflammatory drugs and regional anesthesia, we were able to reduce intraoperative opioid use at our facility by 90% without compromising patient outcomes or value. Research such as a multicenter study involving a broader variety of cases and surgical settings is needed to determine if our quality improvement project is relevant for other institutions. Further investigation is also needed on postoperative pain and opioid consumption after patient discharge.

CONTRIBUTORS

Travis H. Allen, CRNA, and Lloyd P. Provost, MS.

ACKNOWLEDGMENTS

We thank Kayla Reece, Renelle Risley, Henry Bahnson, and all of the physicians, nurse anesthetists, nurses, pharmacists, and perioperative staff who helped with this quality improvement project.

DISCLOSURES

Name: Amber M. Franz, MD, MEng.

Contribution: This author helped write the final version of the manuscript, make substantial contributions to data, cost analyses, manuscript revisions, and approve the final version for publication.

Conflicts of Interest: None.

Name: Lynn D. Martin, MD, MBA.

Contribution: This author helped make substantial contributions to project design, execution, manuscript revisions, cost analyses, and approve the final version for publication.

Conflicts of Interest: L. D. Martin is a shareholder in MDmetrix.

Name: David E. Liston, MD, MPH.

Contribution: This author helped make substantial contributions to project design, execution, manuscript revisions, and approve the final version for publication.

Conflicts of Interest: D. E. Liston is a shareholder in MDmetrix.

Name: Gregory J. Latham, MD.

Contribution: This author helped make substantial contributions to project design, execution, manuscript revisions, and approve the final version for publication.

Conflicts of Interest: None.

Name: Michael J. Richards, BM.

Contribution: This author helped make substantial contributions to project design, execution, manuscript revisions, and approve the final version for publication.

Conflicts of Interest: M. J. Richards is a shareholder in MDmetrix.

Name: Daniel K. Low, BM, BS.

Contribution: This author helped write the first draft of the manuscript, make substantial contributions to project design, execution, and data analyses, and approve the final version for publication.

Conflicts of Interest: D. K. Low is the Chief Medical Officer and founder of MDmetrix, the software program used to analyze the data.

This manuscript was handled by: James A. DiNardo, MD, FAAP.

    REFERENCES

    1. Ogrinc G, Davies L, Goodman D, Batalden P, Davidoff F, Stevens D. SQUIRE 2.0 (Standards for QUality Improvement Reporting Excellence): revised publication guidelines from a detailed consensus process. BMJ Qual Saf. 2016;25:986–992.
    2. Office of the Commissioner. Statement from Douglas Throckmorton, M.D., deputy center director for regulatory programs in FDA’s Center for Drug Evaluation and Research, on the agency’s response to ongoing drug shortages for critical products [Internet]. U.S. Food and Drug Administration. 2018[cited September 30, 2019]. Available at:https://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm611215.htm. Accessed September 30, 2019.
    3. Bruera E. Parenteral opioid shortage — treating pain during the opioid-overdose epidemic [Internet]. N Engl J Med. 2018;379:601–603.
    4. Hollingsworth H, Herndon C. The parenteral opioid shortage: causes and solutions. J Opioid Manag. 2018;14:81–82.
    5. CDC. Drug Overdose Deaths | Drug Overdose | CDC Injury Center [Internet]. 2019[cited September 30, 2019]. Available at: https://www.cdc.gov/drugoverdose/data/statedeaths.html. Accessed September 30, 2019.
    6. CDC. FastStats [Internet]. 2019[cited September 30, 2019]. Available at: https://www.cdc.gov/nchs/fastats/injury.htm. Accessed September 30, 2019.
    7. Brummett CM, Waljee JF, Goesling J, et al. New persistent opioid use after minor and major surgical procedures in US Adults. JAMA Surg. 2017;152:e170504.
    8. Harbaugh CM, Lee JS, Hu HM, et al. Persistent opioid use among pediatric patients after surgery. Pediatrics. 2018;141:e20172439.
    9. Koepke EJ, Manning EL, Miller TE, Ganesh A, Williams DGA, Manning MW. The rising tide of opioid use and abuse: the role of the anesthesiologist. Perioper Med (Lond). 2018;7:16.
    10. Hah JM, Bateman BT, Ratliff J, Curtin C, Sun E. Chronic opioid use after surgery: implications for perioperative management in the face of the opioid epidemic. Anesth Analg. 2017;125:1733–1740.
    11. Kessler ER, Shah M, Gruschkus SK, Raju A. Cost and quality implications of opioid-based postsurgical pain control using administrative claims data from a large health system: opioid-related adverse events and their impact on clinical and economic outcomes. Pharmacotherapy. 2013;33:383–391.
    12. Rees GJ, Gray TC. Methyl-N-propyl ether. Br J Anaesth. 1950;22:83–91.
    13. Chia YY, Liu K, Wang JJ, Kuo MC, Ho ST. Intraoperative high dose fentanyl induces postoperative fentanyl tolerance. Can J Anaesth. 1999;46:872–877.
    14. Mauermann E, Filitz J, Dolder P, Rentsch KM, Bandschapp O, Ruppen W. Does fentanyl lead to opioid-induced hyperalgesia in healthy volunteers? A double-blind, randomized, crossover trial. Anesthesiology. 2016;124:453–463.
    15. Angst MS, Koppert W, Pahl I, Clark DJ, Schmelz M. Short-term infusion of the mu-opioid agonist remifentanil in humans causes hyperalgesia during withdrawal. Pain. 2003;106:49–57.
    16. Fletcher D, Martinez V. Opioid-induced hyperalgesia in patients after surgery: a systematic review and a meta-analysis. Br J Anaesth. 2014;112:991–1004.
    17. Mahmoud M, Mason KP. Dexmedetomidine: review, update, and future considerations of paediatric perioperative and periprocedural applications and limitations. Br J Anaesth. 2015;115:171–182.
    18. Blaudszun G, Lysakowski C, Elia N, Tramèr MR. Effect of perioperative systemic α2 agonists on postoperative morphine consumption and pain intensity: systematic review and meta-analysis of randomized controlled trials. Anesthesiology. 2012;116:1312–1322.
    19. Shariffuddin II, Teoh WH, Wahab S, Wang CY. Effect of single-dose dexmedetomidine on postoperative recovery after ambulatory ureteroscopy and ureteric stenting: a double blind randomized controlled study. BMC Anesthesiol. 2018;18:3.
    20. Franz AM, Dahl JP, Huang H, et al. The development of an opioid sparing anesthesia protocol for pediatric ambulatory tonsillectomy and adenotonsillectomy surgery—a quality improvement project. Pediatr Anaesth. 2019;29:682–689.
    21. Institute of Medicine, Committeeon the Learning Health Care System in America. Best Care at Lower Cost: The Path to Continuously Learning Health Care in America. 2013.Washington, DC: National Academies Press;
    22. Toussaint JS, Berry LL. The promise of Lean in health care. Mayo Clin Proc. 2013;88:74–82.
    23. Langley GJ, Moen RD, Nolan KM, Nolan TW, Norman CL, Provost LP. The Improvement Guide: A Practical Approach to Enhancing Organizational Performance. 2009.San Francisco, CA: John Wiley and Sons;
    24. Wong DL, Baker CM. Pain in children: comparison of assessment scales. Pediatr Nurs. 1988;14:9–17.
    25. Merkel SI, Voepel-Lewis T, Shayevitz JR, Malviya S. The FLACC: a behavioral scale for scoring postoperative pain in young children. Pediatr Nurs. 1997;23:293–297.
    26. Phelps JR, Russell A, Lupa MC, et al. High-dose dexmedetomidine for noninvasive pediatric procedural sedation and discharge readiness. Paediatr Anaesth. 2015;25:877–882.
    27. Aldrete JA. The post-anesthesia recovery score revisited. J Clin Anesth. 1995;7:89–91.
    28. Provost LP, Murray S. The Health Care Data Guide: Learning From Data for Improvement. 2011.San Francisco, CA: John Wiley and Sons;
    29. Vetter TR, Morrice D. Statistical process control: no hits, no runs, no errors? Anesth Analg. 2019;128:374–382.
    30. Frauenknecht J, Kirkham KR, Jacot-Guillarmod A, Albrecht E. Analgesic impact of intra-operative opioids vs. opioid-free anaesthesia: a systematic review and meta-analysis. Anaesthesia. 2019;74:651–662.
    31. Jin S, Liang DD, Chen C, Zhang M, Wang J. Dexmedetomidine prevent postoperative nausea and vomiting on patients during general anesthesia: a PRISMA-compliant meta analysis of randomized controlled trials. Medicine (Baltimore). 2017;96:e5770.
    32. Dexter F, Epstein RH, Rodriguez LI. Throughout the United States, pediatric patients undergoing ambulatory surgery enter the operating room and are discharged earlier in the day than are adults. Perioper Care Oper Room Manag. 2019;16:100076.
    33. Rabbitts JA, Groenewald CB, Moriarty JP, Flick R. Epidemiology of ambulatory anesthesia for children in the United States: 2006 and 1996. Anesth Analg. 2010;111:1011–1015.
    34. Fretheim A, Tomic O. Statistical process control and interrupted time series: a golden opportunity for impact evaluation in quality improvement. BMJ Qual Saf. 2015;24:748–752.
    35. Mascha EJ, Sessler DI. Segmented regression and difference-in-difference methods: assessing the impact of systemic changes in health care. Anesth Analg. 2019;129:618–633.
    36. Manne SL, Jacobsen PB, Redd WH. Assessment of acute pediatric pain: do child self-report, parent ratings, and nurse ratings measure the same phenomenon? Pain. 1992;48:45–52.
    Copyright © 2020 International Anesthesia Research Society