Pediatric Subcommittee Meeting Minutes 10.16.26

Anesthesiology Performance Improvement and Reporting Exchange (ASPIRE)

Pediatric Subcommittee Meeting Minutes – October 16, 2026

Attendance:

Morgan Brown (Boston Children’s)

Michelle Huntington (Corewell West)

Kate Buehler (MPOG)

Cathy Jones (Boston Children’s)

Ellen Choi (UChicago)

Eva Lu-Boettcher (UWisconsin)

Deb Claybaugh (MyMichigan)

Aqsa Sarwar (MPOG)

Joe Cravero (Boston Children’s)

Ashka Shah (Utah Health)

Jurgen de Graaf (NYP-Cornell)

Ruchika Sharma (UVirginia)

Megan Dewey (Indiana)

Brady Still (UChicago)

Kristen Groody (UMichigan)

Brian Waldschmidt (UVermont)

Ruchika Gupta (UMichigan)

Blake Wilson (MyMichigan)

Melanie Herren (MPOG)

Start: 0700 PST (1000 EST)

Minutes from June 01, 2026 meeting approved - minutes, slides and recording posted on the MPOG

website for review

Patient Blood Management and Preoperative Anemia Measure

Discussion

• Morgan Brown (Boston Children’s) provided an update on the patient blood management

metric, which focuses on screening for anemia in patients at high risk of transfusion.

• The target population was built using CPT codes associated with transfusion risk, excluding

emergency surgeries, transplants, burns, cardiac cases, neonates, diagnostic imaging cases, very

short cases, and patients without a documented preoperative hemoglobin.

• The metric was recently published. The final cohort had a transfusion rate above 5%, and about

one-third of patients undergoing major non-cardiac surgery presented anemic — higher than

expected.

• Some centers showed no anemic patients, raising the question of whether this reflects small

sample sizes or differences in practice, such as measuring and deferring surgery.

• Cathy Jones (Boston Children’s) and Joe Cravero (Boston Children’s) noted limitations, including

the lack of known baseline anemia rates for a normal pediatric surgical population and the fact

that most children do not receive preoperative blood work.

Decisions

• This is considered version 1.0 of the metric; no final determination was made on further

refinement.

Next Steps

• Attendees were asked to consult their QI champion, review their institution's data, and provide

feedback on whether the metric is useful and how it could be refined.

Pediatric Pain and Opioid Measure Discussion

Discussion

• Morgan Brown (Boston Children’s) shared Boston Children's PONV data for tonsillectomy,

adenoidectomy, and adenotonsillectomy cases which were consistent with published national

data (the CHEER study, cited at approximately 1.3%).

• The group discussed direction for a new pain and opioid metric, including whether to narrow the

focus to a specific procedure phenotype (tonsils, spine, outpatient orthopedics) rather than

looking across all patients.

• Eva Lu-Boettcher (UWisconsin) emphasized separating inpatient and outpatient populations,

since recovery priorities and protocols (such as ERAS) differ significantly between the two.

• Spine surgery was discussed as a complex candidate due to variable post-op course and

potential ICU involvement, making follow-up difficult; tonsillectomy/adenotonsillectomy or

outpatient orthopedic cases were seen as better procedure options to focus on for V1 of this

measure.

• Cathy Jones (Boston Children’s) referenced a pain-free, same-day protocol presented at a

Society for Pediatric Anesthesia (SPA) quality and safety meeting by a group from the University

of Wisconsin, which may offer comparison data once published.

Decisions

• General agreement that tonsillectomy, tonsillotomy, and adenoidectomy represent an

achievable starting cohort for a pain/opioid metric.

Next Steps

• Continue evaluating which outcome measures (peak pain scores, pain terciles, opioid use,

rescue opioid rates) would be most meaningful to track.

Potential Procedure Cohorts

Discussion

• The group discussed additional candidate procedure types beyond tonsils and adenoids,

including outpatient orthopedics (noted for heavy catheter and block use) and spine surgery

(considered less feasible due to variable post-op courses).

• Brian Waldschmidt (UVermont) noted via chat that it would be valuable to examine whether

intraoperative opioids affect PACU opioid needs and pain scores, and whether specific

multimodal medications (ketamine, precedex, NSAIDs) show a signal.

• Morgan Brown (Boston Children’s) noted that starting with tonsils and adenoids would be easier

from a programming standpoint than other procedure types.

Decisions

• No final cohort was selected; tonsils/adenoids remained the leading candidate, with outpatient

orthopedics identified as a secondary area of interest.

Next Steps

• Determine whether peak pain scores or opioid-based measures (intraoperative stratification,

post-op outcomes) should be prioritized for the new metric.

Opioid Equivalency Measure and Procedure Phenotypes

Discussion

• Morgan Brown (Boston Children’s) announced that the opioid equivalency measure for

tonsillectomy had been pulled back from the dashboard due to calculation issues and proposed

updates by Clark Fisher (Yale) as part of a standard measure review at a recent Quality

Committee meeting.

• The group clarified the distinction between a reported metric and a research project: opioid

equivalency data will be used for research and QI purposes while remaining informational only

on the dashboard, rather than a pass/flag metric.

• Plans include rolling out opioid equivalency calculations for a limited adult procedure group first,

confirming the calculations are accurate, and then extending to pediatrics.

Decisions

• The opioid equivalency measure will remain off the dashboard until calculations are refined, and

will be treated as informational data rather than a scored metric going forward.

Next Steps

• Refine opioid equivalency calculations starting with the adult population before extending to

pediatric procedures (likely spine or another group identified by the subcommittee).

• Identify additional pediatric procedure type phenotypes beyond

tonsillectomy/adenotonsillectomy to support this work.

Metric Development Priorities and Outcomes

Discussion

• Morgan Brown (Boston Children’s) posed two questions to the group: (1) whether

intraoperative details, such as ketamine or precedex use, matter beyond outcomes alone for

adenotonsillectomy, and (2) which outpatient orthopedic procedure (for example, ACL) should

be considered for a new phenotype.

• Cathy Jones (Boston Children’s) and Eva Lu-Boettcher (UWisconsin) discussed the variability

within “outpatient ortho” and whether procedures should be stratified by case type (for

example, ACL versus MPFL reconstruction), given differing recovery patterns and block

techniques.

• Joe Cravero (Boston Children’s) noted the challenge of limited preoperative information and

variability in patient populations (for example, prior opioid use or depression) affecting

outcomes, and expressed interest in examining block use (such as paravertebral blocks) in

cardiac and other phenotypes, noting that data from CHOP is expected soon.

• The group distinguished between quality-improvement-focused metrics and research questions,

noting both could proceed in parallel.

• A question was raised about whether comorbidity filters (such as BMI or AHI/sleep apnea) could

be added to help explain variation in pain scores for tonsillectomy patients; Morgan Brown

(Boston Children’s) noted this could potentially be requested from MPOG for one or two

comorbidities.

Decisions

• No final decision was made on outpatient orthopedic procedure stratification; views differed,

with Cathy Jones (Boston Children’s) favoring a broad outpatient ortho category and Eva Lu-

Boettcher (UWisconsin) favoring stratification by case type (for example, ACL versus MPFL

reconstruction).

Next Steps

• Explore adding comorbidity filters (such as sleep apnea) to MPOG for tonsillectomy-related

metrics, pending feasibility.

• Consider starting with pain scores and/or opioid rescue needs as an initial outcome measure

before expanding to intraoperative technique analysis.

Next Steps and Closing Remarks

Discussion

• Two additional metric updates are needed, and members were asked to consider taking one on;

Morgan Brown (Boston Children’s) noted the process is manageable since metrics are only

updated every couple of years.

• Feedback was requested on TRAN-06 (confirmed to be anemia-only, not broader red

cell/transfusion data) to guide further refinement, with a goal of finalizing direction by fall of

next year.

Decisions

• TRAN-06 will remain focused on anemia only.

Next Steps

• Volunteers are needed for two upcoming metric updates: PONV-04-Peds and FLUID-02-Peds.

• Continue gathering feedback on TRAN-06 to finalize direction by fall of next year.

• Next subcommittee meeting is tentatively planned for February.

Adjourned: 0750 PST (1050 EST)

Full Transcript

Pediatric Subcommittee Meeting Transcript

October 16, 2026

Introductions and Institutional Quality Metrics

Morgan Brown (Boston Children’s): Good morning. Thanks to everyone who's joined us on Zoom. We

have a handful of people here in a very large room. It looks like everything is working now. This format is

designed to help us meet each other and keep things a little more informal, but we will see how it goes.

Kate Buehler (MPOG) is monitoring the chat, so if you have any questions or want to raise your hand

and say something, please do so. I want to start with some names and introductions. Are you part of

MPOG? Do you know who your QI champion is? Are you sending people metrics from MPOG? Are you

using any other metrics at your institution — for example, tracking Zofran use or proper narcotic

reconciliation? What are you using at your institution?

I'll come over here so that everyone can hear.

Ruchika Sharma (UVirginia): I'm from the University of Virginia Children's Hospital in Charlottesville.

Yes, we are part of MPOG. I'm not sure who our QI champion is, but I believe our PI is BKN Nike, an adult

anesthesiologist. We receive the monthly metrics — I believe that is the only thing we have.

Eva Lu-Boettcher (UWisconsin): I'm from the University of Wisconsin. I am the quality champion for

both adult and pediatric anesthesia, and it's a pleasure to be here. We currently don't use any other

metrics. We used to have other metrics, like AQI, but not currently.

Ruchika Gupta (UMichigan): I'm at the University of Michigan. We are part of MPOG and have several

QI champions, though we don't have a pediatric-specific one — Nera is probably our biggest one. We

send our MPOG metrics to all of our faculty, and we don't use any other metrics besides what we call

the Aspire metric.

Jurgen de Graaf (New York Presbyterian): I'm from Cornell Medicine. We are part of NYP, we do a lot of

research and we receive our monthly QI quality metrics. We don't use any other metrics.

Joe Cravero (Boston Children’s): I'm from Boston Children's. I know who our quality champion is. I'll let

Cathy Jones (Boston Children’s) talk about what we send to people.

Cathy Jones (Boston Children’s): I'm also at Boston Children's, and like Joe Cravero (Boston Children’s), I

know our QI champion. In addition to MPOG, we send people metrics if they have narcotic

discrepancies, because we have a new system. For the division I'm in, we send information to people

about medication scanning compliance, and we'll probably have other metrics soon.

Morgan Brown (Boston Children’s): Can we see if anyone online would like to comment first?

[Via Chat] 00:19:07 Brian Waldschmidt (UVermont): Brian Waldschmidt (UVermont), MD, from

University of Vermont. Yes, part of MPOG. Yes, sending metrics from MPOG Aspire to all adult

and pediatric anesthesia providers. No other metrics. Quality Champion at UVM is Paige

Georgiadis, MD (adult anesthesiology).

Morgan Brown (Boston Children’s): Hi, Brian. Do you want to speak? I saw you put a comment in the

chat.

Brian Waldschmidt (UVermont): Hi, yes, I was answering those questions. I'm from the University of

Vermont. I'm the Pediatric MPOG Champion at my institution, and there's an additional quality

champion, Dr. Georgiadis, who is an adult anesthesiologist. Excited to be here.

Morgan Brown (Boston Children’s): Do you use any other quality metrics besides MPOG?

Brian Waldschmidt (UVermont): No, we started sending the Aspire monthly metric emails just over the

summer, so those are our main metrics.

Morgan Brown (Boston Children’s): Melanie Herren (MPOG), are you able to unmute? Deb Claybaugh

(MyMichigan), Michelle Huntington (Corewell West), Blake Wilson (MyMichigan) — would anyone else

like to comment on what they're doing at their institution?

Ellen Choi (UChicago): I don't have particular comments, but I'm Ellen Choi (UChicago), from the

University of Chicago. Yes, we are part of MPOG. We cover both adults and children. Our main champion

is Avery Tung, who is also on the adult side. We send the Aspire notices to everyone.

Morgan Brown (Boston Children’s): Great. Are you using any other metrics in Chicago, Ellen?

Ellen Choi (UChicago): I can't recall at this point — I don't think we have many other metrics that we

routinely send to people.

[Via Chat] Megan Dewey (Indiana): We are new to MPOG at Riley (IU), starting to send MPOG

metrics by the end of 2026. I'm the pediatric champion, and we have an adult-side champion,

Alex Service. No other metrics.

Ellen Choi (UChicago): We do follow data through Epic, and we recently switched over to Omnicell for

all of our medications, so I'm sure there will be reports coming out from there at some point.

Morgan Brown (Boston Children’s): Great — I see Megan at Riley Children's says that she is going to

start sending metrics by the end of this year, and she's the pediatric champion. Would anyone else like

to introduce themselves or comment?

Blake Wilson (MyMichigan): I just wanted to introduce myself — Blake Wilson (MyMichigan), from

Michigan Medical Center. I'm the System Quality and Safety Champion. We don't currently have a

pediatric-specific champion; all of our sites care for both adult and pediatric patients. We're part of

MPOG and Aspire, and we don't submit data to any other consortium.

Morgan Brown (Boston Children’s): Great, nice to meet you, Blake Wilson (MyMichigan). That might be

everybody, though I'm not always able to see everyone on the call. Go ahead, Dr. Cravero (Boston

Children’s).

[Via Chat] Megan Dewey (Indiana): Only with the mic!

Morgan Brown (Boston Children’s): Dr. Cravero (Boston Children’s) was asking for feedback on whether

people are using these metrics for anything beyond general interest — are they being used for

evaluation or anything else at your institution? Thank you, everyone, for that. We do use it at Boston

Children's; it's part of the yearly faculty discussion with the division chief, although we don't really know

how much people are discussing it. That's why we're asking — what are people really using this for?

Patient Blood Management and Preoperative Anemia Measure

Morgan Brown (Boston Children’s): I want to update everyone on where we are with the patient blood

management metric we've been discussing for the last year or so. As you may recall, we looked at the

literature on patient blood management, and we decided to focus on the first recommended step —

screening for anemia in patients who are at high risk of transfusion. After several meetings, we worked

on defining our target population.

That turned out to be more difficult and complicated than expected, as is often the case when working

with this much data. When you look at the guidelines, the point of treating anemia is that you might be

able to optimize these patients rather than cancel or delay surgery. We first wanted to identify a group

of patients at risk of transfusion, because it doesn't make much sense to focus on anemia if a patient

isn't going to be transfused. We started by looking at CPT codes for who did and didn't get transfused,

and came up with a list that is available on the MPOG definitions site if you'd like to review it. We used

those specific CPT codes and excluded several other groups.

We excluded emergency surgery patients, since there wouldn't be an opportunity to treat anemia

beforehand. We excluded transplants and burns, since even if those cases aren't emergent, it's unlikely

surgery could be deferred for any meaningful length of time. We excluded cardiac cases because they

introduce a lot of additional complexity. We excluded neonates, since if a neonate is coming in for

surgery, there is usually a reason for it, and deferring for six weeks is generally not reasonable. We

removed diagnostic imaging cases and very short cases, since those likely weren't large procedures, and

we excluded patients without a preoperative hemoglobin.

I don't have a lot of data to share today, because Meredith is on maternity leave, so I'm working with

what I have access to. This metric was published just a few weeks ago. That's part of why I wanted to

see what people have access to, since your quality champions can see more than you can see as an

individual. This is one of those metrics we looked at from an institutional standpoint rather than an

individual one, because most individuals wouldn't have enough cases to make it meaningful.

At any rate, it has been published, and I'm hoping people will look at it and give us feedback on how we

should refine it further, because it is genuinely tricky to identify the right population. The good news is

that when we looked at the final group, they had a high transfusion rate — over 5% of those patients

were transfused. Blue represents Boston, because the dashboard shows which institution you're

viewing. This is one of those metrics where a higher number is actually better.

About a third of our patients having major non-cardiac surgery come in anemic, which is higher than I

expected. Some centers didn't have any patients coming in anemic, and I think we need to look further

into whether that reflects small sample sizes or whether those centers are measuring and deferring

surgery. At any rate, once we move forward — sorry, this is my Vanna White moment.

Cathy Jones (Boston Children’s): Even for our own data, I don't know what our baseline anemia rate is

in our normal population. I have no idea whether this is an average for anyone presenting for surgery or

whether there's something specific about this population, and I don't know if MPOG could determine

that either.

Morgan Brown (Boston Children’s): One factor is that most children don't get preoperative blood work,

so for most low-risk surgery, they won't have a hemoglobin on file. It's a little startling. We also haven't

shown here how severe the anemia is, since some patients could be only mildly anemic, and there is

debate about what hemoglobin threshold should define anemia.

Joe Cravero (Boston Children’s): I would only add that anemia rates in children at various ages are

known for the general population. We may not know the rate for our exact population, but anemia rates

among one-year-olds and two-year-olds are known numbers if you wanted to compare to the general

population. This is our first attempt at patient blood management, which is genuinely challenging and

probably why it hasn't been done before.

Morgan Brown (Boston Children’s): Our request is for you to find your QI champion, look at your own

institution's data, and see what you think of this metric. Is it useful? Does it still need to be refined to be

more useful? We can absolutely continue working on this — this is just version 1.0. Does anyone online

have questions or comments?

Pediatric Pain and Opioid Measure Discussion

Morgan Brown (Boston Children’s): This is Boston Children's data, but again, it's what I was able to pull

together. I'm showing you an example of a metric that looks at a wide variety of procedures, which is

how many of the metrics work — looking at PONV or temperature across almost everybody. As a QI

champion, you can also use filters I added to the dashboard to look at specific procedure types. Only a

few are available right now, and one of them is tonsillectomy, adenoidectomy, and adenotonsillectomy.

I selected those procedure types and excluded plain adenoidectomy-only cases, then looked just at

those patients. If you were at an institution that wanted to target this population, MPOG can already

show you how many cases you've done — in the last year, we did over 2,000 of these cases. You can see

that everyone in Boston is getting multimodal analgesia, as described by that metric.

A study called CHEER, published a few years ago, reported a rate around 1.3%, with a fairly wide

confidence interval. We don't currently have anything in MPOG about pain scores or opioid use.

Morgan Brown (Boston Children’s): Fellows and residents are included, which is why there are so many

providers attached to those cases, not just attendings.

That brings up the point I wanted to spend most of today's discussion on: what direction do we want to

take for pain and opioid measures — where should the next metric go? I want to leave it open for

everyone to discuss what would matter most to your institution, for clinical care and for individual

performance assessment. Should we narrow down to a specific phenotype that already exists rather

than looking at all patients — tonsils, spine patients, some other group? Once we decide that, what do

people feel is the most representative thing to look at, understanding that every option has limitations

— peak pain scores, pain terciles, total opioid use in the recovery room, percentage of patients requiring

rescue opioids in the recovery room? There's a lot to consider.

I'm not a pain expert, but I know there are people here who are, so I'd like to hear your thoughts on

what would be helpful. Eva Lu-Boettcher (UWisconsin), would you like to start?

Eva Lu-Boettcher (UWisconsin): I think if we're going to build a pain-based metric, it would be helpful to

separate the inpatient and outpatient populations, because as a pain physician I view those patients

very differently. A lot of our inpatients have some kind of ERAS protocol, while outpatient procedures

prioritize different things. For inpatients, I'd look at rescue medications given, though I'm not sure how

good the post-op capture window is. As a pain physician, I care a lot more about the overall inpatient

stay experience compared with outpatient care. These metrics could be built, but they would look very

different depending on the patient population.

I'll state the obvious — spine cases are interesting, but without inpatient data on how much opioid

patients receive postoperatively, it wouldn't help much. I'd also note that spine patients are a complex

group unless you really subdefine them — for example, adolescents with idiopathic scoliosis and no

other underlying medical or psychological conditions are a somewhat narrower group.

Personally, what interests me is that there's a lot of thought, particularly from the group in Seattle, that

opioid-free anesthesia for outpatients, especially tonsillectomy and adenoidectomy, is the way to go. I

think it would be interesting to use MPOG's ability to look at a large group of children having these

operations and subdivide them into groups receiving a significant amount of opioids, a small amount, or

none at all. Since we don't have post-discharge or post-PACU data if we're not admitting, I think post-op

information would be very interesting, but we could at least look at the PACU to see if there's a real

signal of superiority for any of those groups, including no opioids at all — rate of rescue opioid use in the

PACU, earlier discharge, less nausea and vomiting, and so on.

With a large enough group, I think this is still an unsettled topic, even though those who strongly favor

an opioid-free approach might consider it settled. I'm not sure that's the case. I would suggest looking

specifically at tonsillectomy, tonsillotomy, and adenoidectomy.

Eva Lu-Boettcher (UWisconsin): I think we'd all agree there are some subtle differences between those

two — our ENT surgeons will say tonsillotomy is quite different. I'm not sure I fully agree, but I think

you'd need to subdivide to understand that difference. I'm just raising it as a consideration.

The other thing we don't account for is that our surgeons tend to have a fairly fixed discharge time —

they usually want patients to stay in-house for a set period to monitor for bleeding unless something

extraordinary happens. Discharge time specifically for tonsils is hard to interpret for that reason; you can

really only look at outliers, because the typical protocol is to keep patients for about two hours to

monitor for bleeding. So discharge time for that specific surgery is difficult to use as a metric because of

institutional protocols.

I think building a broad pain metric across all surgeries would be next to impossible because of the

number of variables involved.

Cathy Jones (Boston Children’s): That's what I was thinking about too. I don't know if others were at the

quality and safety meeting for SPA, where an anesthesiologist from Delaware talked about joining the

Adaptix group along with a group from the University of Wisconsin — I believe they're planning to

publish soon. They introduced a pain-free, same-day protocol and shared some of their data. It might be

useful to see similar outcomes using MPOG data, depending on what they publish.

Potential Procedure Cohorts

Morgan Brown (Boston Children’s): It sounds like everyone agrees it would make sense to focus on a

specific cohort. The nice thing about tonsils and adenoids is that it's an achievable cohort — there's

already some groundwork done, as shown by this data, even if it isn't perfect. Is there any other surgery

type that would work? Spine doesn't sound like a good fit — is there anything else you can think of?

Eva Lu-Boettcher (UWisconsin): Outpatient orthopedics as well, since our surgeons are very catheter-

and block-heavy. Having a good metric for the ortho group, which makes up the majority of our acute

pain service, would also be valuable. I do think tonsils would be a great starting point.

Spine cases are tricky because so much depends on the post-op course, including whether patients are

on methadone, and whether they go to the ICU — that makes follow-up very difficult. Outpatient

orthopedics, like on the adult side, is something that's very easy to track and meaningful for the

surgeons as well. I'm not sure how feasible it would be to track ED bounce-backs, since we see very few

of those due to our follow-up process for catheters and single-shot blocks, but it's something to

consider.

[Via Chat] Brian Waldschmidt (UVermont): Agree — would be most valuable to look at the

tonsillectomy/adenotonsillectomy population for whether intraoperative opioids are given and

whether that impacts whether PACU opioids are needed (and possibly PACU pain scores). Might

even find a signal for which multimodal medication had impact (ketamine, precedex, NSAIDs).

Morgan Brown (Boston Children’s): As many of you know, I do cardiac anesthesia, so I don't have a

strong preference about which specific item we pick here. The nice thing about starting with tonsils and

adenoids is that it's easier than most other options, at least in terms of programming time. It may be

possible to use different filters to look at a few different things within this cohort. Do you think peak

pain scores are most important? That's what the Seattle group published — I read their opioid-free

paper a couple of years ago, and they focused on peak score. I know Joe Cravero (Boston Children’s) has

previously advocated for more of a tercile-based approach — high, medium, and low pain. Alternatively,

do you think we should focus more on opioids, and if so, stratifying what patients received

intraoperatively and then tracking what happens postoperatively? What do you think would be most

useful, both individually and at the institutional level?

Cathy Jones (Boston Children’s): I'd offer that recovery is a composite outcome, so you can't look at just

one thing. The issues that matter most to most of us are pain level, nausea and vomiting, agitation (if it's

recorded), and duration of recovery time. On that point — agreeing with the duration issue — if the

duration of stay is fixed, does it matter if the patient wakes up a little quicker? That raises a lot of

different questions.

For me, it would be interesting to look at the techniques used — opioids, dexmedetomidine, ketamine,

whatever people are using — and then look at composite outcomes. Highlighting an age group and a

surgery type we can all agree on, and looking at the variation in technique alongside those composite

outcome metrics, could show whether one particular approach yields a better composite outcome than

another.

There's probably a lot of discussion needed about what exactly that should look like, but I don't think

there's a convincing case yet for an absolute advantage to a completely opioid-free approach — though

people could certainly argue that point. I think this group could do something genuinely interesting

here.

Morgan Brown (Boston Children’s): That's being worked on right now for adults, so I think it's being

addressed, but we weren't planning to tackle it immediately.

Opioid Equivalency Measure and Procedure Phenotypes

Kate Buehler (MPOG): Sorry, I'll repeat myself. We just retired, or pulled back, the opioid equivalency

measure for tonsillectomy from the dashboard because there are some calculation issues on the back

end, and there's been a lot of debate. Clark Fisher from Yale recently reviewed this measure at a quality

committee meeting and raised some good points about whether we should include neuraxial anesthesia

when calculating opioids. There isn't good literature to tell us what those calculations should look like,

so we're not sure whether neuraxial should be included or excluded. Oral morphine equivalents are

fairly straightforward, and there's a reasonable amount of literature on converting between oral and IV

routes, but not for many of the other routes. So we're considering pulling back a lot of that calculation

until there's literature to support it.

As we do this, we plan to roll the measure out for a limited procedure group, starting with adults, to

make sure the math holds up, then roll it out for pediatrics if it proves meaningful — likely spine or

whatever the group felt was most useful. We're also working on new procedure type phenotypes at the

same time, so it would help to know if there's something besides tonsillectomy/adenotonsillectomy that

would be useful from a pediatric perspective. We can start working on those new phenotypes while we

finish ironing out the opioid equivalency calculations for adults. This would be a good time to suggest

procedure types you'd be interested in if tonsillectomy isn't the focus area.

Morgan Brown (Boston Children’s): I haven't read the paper Dr. Fisher is working on, or whether it's

published yet, but he's looking at some interesting concepts — for example, giving a patient all of their

opioid at the beginning of a case is different from giving opioid throughout the case, and he's exploring

curves and modeling techniques to determine how much opioid a patient really received, since he's

approaching it from a research perspective. He presented some of that work when looking at

intraoperative opioid use, including how to account for remifentanil — does it count at all, since it's

often considered free? Even though those equivalency tables are well published, I think some of the

values may not be well substantiated — I haven't been able to trace where some of the numbers

originated.

Joe Cravero (Boston Children’s): We're talking about two different things here — does this become a

reported metric, or is this a research project?

Morgan Brown (Boston Children’s): If we're looking at this from a research perspective, we might take

an age group of children having a particular procedure and do a deep dive into exactly what they're

receiving in terms of opioids. The nice part about tonsillectomy and tonsillotomy is that nobody is

getting a spinal for those cases, and with an operative time of 10 to 12 minutes, the timing of opioid

delivery is probably less important than it would be for a three-hour operation. There's a lot to discuss,

but I want to draw a clear distinction.

Cathy Jones (Boston Children’s): I believe they were originally treating it as a metric, but after the

quality committee meeting, they decided to take it off and fix some issues before it returns as a quality

metric. We never really used it at Boston, because it looked a little off to me and I couldn't quite figure it

out, but that's exactly right — we had a metric around this, not many people used or looked at it, and

when people started examining it, they found issues, so it was rolled back.

Morgan Brown (Boston Children’s): We need to refine the calculations behind the scenes, but research

is happening at the same time — these efforts are running in parallel. What we're focusing on right now

is purely informational, not a metric in the same way as the rest of the MPOG measures, which have

pass and flag thresholds. This is purely informational: here is what the opioid equivalency was for your

case, and if you line that up with outcomes, it could be meaningful. That's what we're thinking of doing.

Unidentified Speaker: Wasn't there also a separate measure about narcotic use — the one that showed

the average amount of narcotic used during a case, where you could select different case types? It was

interesting to look at, though it wasn't always clear what to do with the information. A few of the

numbers seemed off, and we didn't get much further with it. In the last couple of months, they took a

closer look and decided to use the available research to refine it rather than leave it as is.

Metric Development Priorities and Outcomes

Morgan Brown (Boston Children’s): To bring us back to this a bit — I think there are two questions I'd

like opinions on. First, beyond just looking at outcomes, is there interest in knowing what's being done

intraoperatively for adenotonsillectomies? Is that information useful — does ketamine use matter, does

precedex use matter — or is it only the outcomes that matter? Second, regarding outpatient

orthopedics, since Kate Buehler (MPOG) mentioned there's an opportunity to develop another

phenotype, is there a specific procedure — ACLs, for example — that you'd want to focus on?

Eva Lu-Boettcher (UWisconsin): If a place is having better outcomes with their cases, I want to know

how they're getting there. Just knowing they have better outcomes isn't particularly useful if we can't

replicate it, but if we can say a site has better outcomes and identify what they're doing differently, that

could be genuinely helpful for patients.

Morgan Brown (Boston Children’s): I agree — without knowing how to get there, just knowing

outcomes differ isn't very actionable. The first step would be looking at pain scores and/or opioid rescue

needs, and if we see meaningful differences, then doing a deeper dive into what's happening. What are

people's thoughts on other procedure types where pain may be a major consideration?

Joe Cravero (Boston Children’s): I do think we'd be interested in a cardiac phenotype, even though right

now we only have six hours of post-op data. Some sites are starting to capture longer periods of in-

hospital data, which should give us more flexibility to look at ERAS pathways and similar questions — a

whole other frontier for both quality improvement and research opportunities. Any thoughts on other

surgeries, like whether ACL is a good target?

Cathy Jones (Boston Children’s): I'd stick with outpatient ortho broadly, because there's a lot of

variability by surgeon in terms of what types of ortho cases they're doing. In general, I think outpatient

ortho of some kind is useful — if the surgeon tells me the patient is going home the next day, I'm more

aggressive with things like single-shot blocks, versus if they're staying one or two days, I have more

room to work with. As a pain team, I think it would be valuable to look at the overall category rather

than picking one specific procedure, since a lot of outpatient ortho procedures are fairly similar in terms

of expectations.

Eva Lu-Boettcher (UWisconsin): Sometimes they're joint procedures, sometimes they're scopes — fairly

similar in that sense. Some patients are discharged home after more invasive procedures like growth rod

placement, and those are still counted as outpatient ortho. I'd be interested in looking at all of those.

Kate Buehler (MPOG): Would surgical service of ortho be a sufficient category on its own?

Eva Lu-Boettcher (UWisconsin): Personally, I think there are too many differences — an ACL

reconstruction and an MPFL reconstruction with a BTB graft, for example, are not the same in terms of

how patients recover, and we use different blocks depending on what's done. I think you should stratify

by case type if you're focusing on outpatient ortho.

Morgan Brown (Boston Children’s): Maybe a way to approach it would be to look at what the most

frequent outpatient orthopedic procedure is and focus there, because one of the problems we run into

with all of our metrics is that it's great to have informational metrics, but we also want to give direct

feedback to providers. In pediatrics, we're often not doing the same procedure frequently enough as

any individual provider to make that feedback meaningful — everything is combined across procedure

types.

Cathy Jones (Boston Children’s): That's right, and procedures do get combined together — we have tiny

infants and 110-kilogram 18-year-olds all grouped in, which makes it hard. When I look at my own

quality report, I sometimes understand why a patient is flagged, but it's not always clear.

Joe Cravero (Boston Children’s): I'd ask how much preoperative information we have, because one issue

we've seen in the Society for Pediatric Anesthesia Improvement Network's work on ACL cases is that

there is a wide range of patients — some are already on opioids coming in, some have significant

depression, and I think outcomes are, to some degree, predicated on preoperative factors. That might

be worth considering. It would be important to determine how to handle a short-term outcome

measure for something that has days or weeks of recovery associated with it. That's going to be a tricky

one, though you could look specifically at PACU outcomes for patients who received blocks versus those

who didn't.

I think the interesting cardiac situation is the blocks that some groups, including ours and the Stanford

group, are doing. If you could isolate a relatively constrained age group and surgery type for patients

receiving paravertebral or other blocks versus centers that aren't doing that, you might find some

interesting information even within the first six hours. I wouldn't eliminate that possibility, especially

where specific interventions are concentrated at certain sites. We're also going to start getting data

from CHOP, which should add a substantial number of patients similar to the ones you're caring for.

Morgan Brown (Boston Children’s): That's a good point, and for all of you — since I know I see many of

you at different venues — it would be great to do more research alongside this work to support these

metrics further, because that's where I think we can have a real advantage.

Joe Cravero (Boston Children’s): I'd also say that if we have enough people involved, there's a quality

metric around whether providers are doing nerve blocks, and whether there are any complications. It's

also a quality question in terms of pain scores across large groups of patients. From a quality

perspective, I'd want to know whether our patients are doing better than the larger group because

we're doing blocks, or whether they're not.

Morgan Brown (Boston Children’s): If I were to approach this, assuming we only have six hours of post-

op data right now, you could look at narcotic use in the first six hours and use that to assess whether

blocks are making any difference at a larger scale. We're working on that in-house — nothing about it is

simple.

[Via Chat] Brian Waldschmidt (UVermont): agree that it would be most valuable to look at the

tonsillectomy/adenoidectomy population for whether intraoperative opioids are given and

whether that impacts whether PACU opioids are needed, and possibly PACU pain scores — and

that we might find a signal for which multimodal medication had impact, such as ketamine,

precedex, and NSAIDs.

Eva Lu-Boettcher (UWisconsin): Related to that comment — for tonsils, is there a way to see whether

patients have a high BMI or a high AHI, to see if there's something different about patients with higher

pain scores? Maybe providers gave less opioid because of comorbidities. Is that something we can see

currently?

Morgan Brown (Boston Children’s): We could potentially ask MPOG about adding that as a filter —

probably not for everything, but maybe one or two comorbidities, since I agree that if a patient has

sleep apnea, you're likely to reduce the dose. That's something people are always asking us for, but we

have to build things that serve multiple purposes to keep the computing burden manageable, since

every new addition takes significant effort — the blood management metric was a big undertaking,

because it was entirely new and different.

Next Steps and Closing Remarks

Morgan Brown (Boston Children’s): Hopefully the next one will be a little simpler. I'll leave it there for

now. I want to make sure everyone has time to get to the rest of the meeting as well. Meredith is still

out, and we may add another quality person at some point before the new year. I'm hoping our next

meeting will be in February. We really need to complete two more metric updates — please consider

taking one on. They're not too bad; I did one for pain, and it wasn't overly complicated, since we only

update them every couple of years, so you only need to review a couple of years of data.

Again, we're looking for feedback on Transfusion 06 and where you think we should make modifications

to make it more useful. Hopefully we can come back with more data, at least on tonsils, and make more

final decisions about the direction of that metric so that we have it ready by fall of next year.

Morgan Brown (Boston Children’s): TRAN-06 only measures anemia. I'm hoping that, longer term, the

phenotype we've built around major non-cardiac surgery could be useful for other purposes we might

want to look at, such as major inpatient issues. As a reminder, everyone is welcome to join the

committee. If you know people who are interested, please have them reach out. We're a small group

within the larger pediatric community and are happy to have more participants. As you know, Eva Lu-

Boettcher (UWisconsin) is our vice chair, so feel free to email either of us with questions. Otherwise,

thank you, and enjoy the rest of the meeting. Thank you to everyone who joined us online.

[Via Chat] Kate Buehler (MPOG): Thank you for joining!