Development of a Structured Regional Analgesia Program for Postoperative Pain Management

We conducted this study in the Nationwide Children’s Hospital (NCH) NICU, which consists of 130 beds in an all-referral children’s hospital. A postoperative multimodal pain management guideline is used and includes scheduled acetaminophen or ketorolac, with either morphine or fentanyl administered as a continuous infusion for procedures deemed to have the potential to cause severe pain. Postoperative pain plans are written by the pharmacist in the medical record for all surgical procedures on the day before surgery to facilitate the medications to be ordered and available at the bedside with intravenous (IV) lines primed before the patients return from surgery to avoid delays in analgesia delivery. Before this initiative, RA for postoperative pain management was used frequently for patients across NCH but was not routinely used for patients in the NICU.^16,17  Our RA team/pain team is comprised of anesthesiologists skilled in pediatric RA. The pain team followed previously established hospital practices for epidural insertion in this population. Our anesthesiologists largely use the technique of cephalad catheter threading through the sacral hiatus, with ultrasound-based confirmation of catheter tip location. To mitigate risk and concerns about local anesthetic systemic toxicity, the pain team avoids amino-amide local anesthetics that rely on immature microsomal enzymes for metabolism in infants. In this population, the continuous epidural infusions contain the amino-ester 2-chloroprocaine at 1.5% concentration with clonidine as adjuvant.

Our primary outcome measure was the percent of eligible surgical procedures that received RA.  Secondary outcome measures included postoperative opioid use and time to extubation. Postoperative opioid use was defined as total opioid exposure (continuous infusion + doses as needed) in the postoperative period in milligrams per kilogram of IV morphine milligram equivalents (MMEs). To account for baseline opioid use before surgery, we tabulated the total postoperative opioid consumption until the time the patient returned to preoperative opioid requirements. The time to extubation was defined as the time needed to extubate after the completion of the surgery for those infants who were not on mechanical ventilator support before the procedure. We tracked the average time to place the epidural catheter and the average duration of epidural use as additional measures to evaluate the functioning of our process. Balancing measures included the Neonatal Pain Agitation and Sedation Scale¹⁸  scores for 48 hours in the postoperative period. We also monitored for catheter-related complications, including dislodgement, early removal, leaking that impairs dressing integrity, occlusion, bleeding, or infection. Given the additional time needed for epidural placement in the operating room (OR), we used hypothermia, defined as a temperature of ≤36.0°C at the time of return to the NICU, as an additional balancing measure.

The study was reviewed and classified as a QI initiative and not human subjects research by the Institutional Review Board at NCH.

We plotted postoperative opioid exposure dosage data in a time series with a statistical process control chart, specifically an I-chart, using a Microsoft Excel plug-in coded by our institution’s statistician. We used standard rules to identify special-cause variation.^19,20  Differences in median opioid use and duration of ventilation were analyzed by using the Kruskal-Wallis test, and proportions of extubated infants and demographics were analyzed by using Fisher’s exact test, using Minitab.²¹ 

Based on the current available evidence of safety and the potential to reduce opioid exposure, we sought to establish a collaborative initiative of neonatology, anesthesia, and surgery in a multidisciplinary model to increase the usage of RA in our NICU for patients undergoing invasive surgical procedures. Our key driver diagram depicts our aim, key drivers, and planned interventions (Fig 1). 

The identification of barriers and changes to current processes for each medical discipline to implement a NICU RA service resulted in the development of an extensive swim lane process map (Fig 2). The map illustrates the multidisciplinary team roles in 3 time phases: presurgery scheduling, day of surgery, and postsurgery. The presurgery time frame displays the structure and collaboration required among 3 departments to determine RA eligibility, obtain parental consent, and schedule OR time for catheter placement. The operative time frame on the day of surgery clarifies the postoperative pain plan responsibility (pain team or neonatology), which is dependent on the successful placement of the epidural catheter. During the postoperative phase of care, the pain team determines when the catheter will be removed, as well as address catheter concerns from the NICU staff.

Patients eligible for RA were defined as infants with a weight of ≥2 kg, undergoing open thoracoabdominal, genitourinary, or extremity procedures, which were scheduled >24 hours in advance and with parental consent. Patients with planned laparoscopic or thoracoscopic operations, known coagulopathies and thrombocytopenia, suspected infections, known spinal anomalies, intraventricular hemorrhage or a ventricular access device, or known neurologic disorders were not considered eligible.

To build support for this project, we held multiple didactic, question-and-answer sessions with surgeons, nurses, nurse practitioners, and neonatologists regarding the benefits and safety of RA. We addressed each of these groups multiple times throughout this initiative and placed education binders, information sheets, and epidural pump cards in strategic areas in the NICU for reference. The pain team helped the nursing staff set up the epidural infusion and epidural documentation in the medical records.

To factor the increased OR time required for catheter placement, an additional 30 minutes was added into OR scheduling with a Current Procedural Terminology (CPT) code for epidural insertion in addition to the surgical CPT code(s) at the time of the surgical case request. This additional built-in time provided assurances that surgeons would not be penalized for having longer-than-usual OR times for the planned surgery.

One of the initial challenges was the determination of who (neonatologist versus surgeon) was responsible for identifying patients for RA. To streamline this process, we worked toward making RA a standard of care, and patients who met criteria had automatic consultations, followed by the pain team determining final candidacy. Our culture has evolved to having RA as the default preference for eligible procedures. The surgeons, pharmacists, and nurse practitioners evaluated OR schedules to determine eligible candidates. This process was fraught with inefficiency and was time consuming. We subsequently developed an electronic, NICU-specific surgery list generated within our electronic medical record that includes the procedure codes as well as the RA CPT code. Using this list not only helps us identify the patients, but it also acts as an audit tool to ensure that we do not miss eligible patients.

NICU began placing pain team consults at least 2 days before the scheduled day of surgery. After we determined the eligibility of the candidates, the pain team obtained parental consent. As the process evolved, we now place the pain team consult at the time of surgical case request. This advanced planning allowed the RA service to be immediately ready for epidural insertion after the primary anesthesia team had finished with induction and line placements.

For the patients who received the epidural catheter, we identified a need for role clarification. The pain team would manage opioid infusions when used for analgesia while the epidural was in place. The neonatology team would resume management of the systemic infusion if used for sedation. Daily pain team rounds provided an opportunity for structured, daily communication and comanagement of pain in a collaborative manner. For the patients who did not receive RA, the neonatology team provided the postoperative analgesia.

After receiving a NICU consultation, the pain team would evaluate and examine the patient for RA eligibility and obtain parental consent if eligible. During the initial stages of the program, there was an occasional communication gap between the NICU and the pain team regarding final determination of eligibility for RA. We resolved this uncertainty by establishing a standardized note written by the pain team’s attending after the initial consult.

Before this initiative, subcutaneous tunneling of the epidural catheters occurred sporadically on the basis of the pain team physician’s preference. The pain team leadership advocated for subcutaneous tunneling of all epidural catheters laterally away from the caudal insertion site to prevent local anesthetic leakage and contamination of the catheter. Tunneling of the epidural catheter was performed by using a technique previously described by Farid et al.²² 

We initiated the project in June 2018. There was a total of 21 patients in the baseline period from December 2017 to May 2018. The baseline data included all patients undergoing similar open thoracoabdominal procedures and meeting other eligibility criteria. After initiation of the project, 34 surgeries met eligibility criteria. Of these, 88% (30 of 34) of infants with eligible surgeries received RA. In the intervention period, epidural catheters could not be placed because of technical difficulty in 2 patients, and 2 other patients were not evaluated for RA by using our new process. Characteristics of the baseline and the intervention group are shown in Table 1; infants in the intervention group were older and had a higher weight at the time of surgery.

The control chart (I-chart) of the postoperative opioid use in patients who received RA is shown in Fig 3. The average postoperative opioid dose decreased from a baseline of 5.0 mg/kg MMEs to 1.1 mg/kg MMEs in the intervention group. The median MME was 3.2 with an interquartile range (IQR) of 1.2 to 7.4 in the baseline group and 0.77 with an IQR of 0.41 to 3.8 in the intervention group (P = .009).

The average time to extubation in the baseline period was 45 hours, as compared to 19.9 hours in the intervention group (n = 19 [baseline]; n = 28 [intervention period]). The box plot of the time to extubation with the median and IQR of the postoperative ventilation hours is shown in Fig 4. During the intervention period, 21 of 28 (75%) infants were extubated in the OR, as compared with 2 of 19 (10.5%) infants in the baseline period (P < .001). The run chart of the median postoperative duration of ventilation hours between the baseline and the intervention groups is shown in Supplemental Fig 5. The average time to place the epidural catheter was 21.7 minutes (SD 9.5 minutes). The average duration of epidural use was 83 hours (range 16–238 hours).

After return to the NICU, 14.3% (3 of 21) of infants in the baseline group and 14.7% (5 of 34) in the intervention group (P = 1.0) were hypothermic. The average Neonatal Pain Agitation and Sedation Scale for the first 48 hours in each individual patient was <4 in both periods. Four out of 30 patients had early catheter removal (2 stooled on dressing, 1 had leakage, and 1 was perceived to be not working). We did not see any other epidural complications, such as catheter dislodgement, infection, systemic seizures, or arrhythmias.

We describe our collaborative efforts in establishing a multidisciplinary program using QI methodologies to increase the usage of RA for postoperative pain management in appropriate patients and in appropriate conditions. With strong, well-developed processes, standardized indications for RA, and extensive education at all levels, we were able to demonstrate a significant reduction in systemic opioid requirements. The majority of patients who received RA were able to be managed with <1 mg/kg IV MMEs, below the threshold associated with adverse neurodevelopmental scores in former extremely low birth weight infants.^8,23  In addition, we showed significant decreases in the duration of postoperative mechanical ventilation, with 75% of patients extubated successfully in the OR postoperatively. The average time to place the epidural catheters fell within the 30-minute period that was predetermined by the program leadership as a reasonable time frame for insertion. Our average catheter-use time of 83 hours approximates indwelling times reported in this population by a large multi-institutional database.¹⁶  Our patients with epidural catheters experienced minimal complications and no serious adverse events, as supported by description in the literature.^15,16 

Developing the process map was critical in helping the team to identify, characterize, and address multiple potential barriers, including when and how to identify eligible patients; to ensure adequate time available in the OR for placing a regional catheter; and for logistics regarding management of patients in the immediate postoperative pain period. Physician leadership was key in facilitating communication among medical specialties and implementing subsequent process change. Extensive education was needed across disciplines to achieve buy-in.

The results of our study are largely supported in the literature. RA is increasingly being used as a part of multimodal postoperative pain management in children and infants,²⁴  and its use has demonstrated opioid sparing, effective analgesia with additional benefits that include decreased time to tracheal extubation and earlier time to full feeds^11–14  as well as shorter ICU stays, and earlier return of bowel function.^{16,17,25–34}  Reassuringly, RA has not been associated with any evidence of long-term neurocognitive impairments in newborn animal models or human infants.^35,36  Postoperative epidural analgesia has also been described in a systematic review as one of the components contributing to enhanced recovery after surgery in pediatric patients.³⁷ 

As the indications for RA continue to expand and the familiarity and skill with the technique grows, this strategy for analgesia is increasing in infants.^16,26,28  In previous studies, authors have described its feasibility and implementation strategy in the pediatric population.^16,26,28  RA is an invasive analgesic technique. Expanding its use in infants to derive the benefits and safety of RA will require the appropriate selection of patients, adequate skills, resources, and training for catheter placement and use.

We acknowledge a number of limitations in our study. We restricted this project to procedures scheduled >24 hours in advance and not urgent or emergent surgical procedures. The underlying diagnoses leading to scheduled versus urgent surgeries may differ; therefore, it is possible that epidural efficacy may differ among underlying diagnoses. However, by standardizing eligibility criteria, we were able to capture a wide variety of thoracic and abdominal surgical diagnoses. So far, we have not evaluated our data relative to surgical site and therefore cannot comment on differential efficacy by location. Duran et al³⁸  showed that laparotomy patients required the addition of an opioid infusion significantly more frequently than thoracotomy patients (67% vs 24%; P = .003). As our next steps, we hope to analyze the data on the basis of the type of surgery and develop criteria and processes for cases booked within 24 hours. In addition, we hope to evaluate measures, such as initiation of first enteral feeds as a result of earlier return of bowel function, increase the use of nonpharmacologic comfort measures such as skin-to-skin care,³⁹  and evaluate factors such as length of stay, and thereby an impact on cost, as our next steps.

We have demonstrated that epidural catheter placement in appropriate patients to provide postoperative RA can reduce postoperative opioid dosage and duration of postoperative ventilation in infants. Multidisciplinary collaboration and multiple process changes to develop appropriate conditions were required to help convert the use of RA from an optional modality to our new standard of care. Our findings further add to the literature supporting the safe and effective use of RA in infants, and provide a framework for other institutions seeking to implement a similar program.

We are grateful for the support and contributions of our anesthesiology team (Lance M. Relland, MD, PhD; Ralph J. Beltran, MD; and Joshua C. Uffman, MD, MBA), nurse educator (Kim Samson, RN), and information systems and QI support staff (Ron Sohner and Venita Robinson, MHSA, CCRP).

CPT

Current Procedural Terminology

IQR

interquartile range

IV

intravenous

MME

morphine milligram equivalents

NCH

Nationwide Children’s Hospital

OR

operating room

QI

quality improvement

RA

regional analgesia