Obstetric Subcommittee Minutes 7.15.20

Obstetric Anesthesia Subcommittee Minutes

July 15, 2020

1:00-2:00pm EST

Zoom

Sharon Abramovitz, Weill Cornell

Angel Martino, Sparrow Health System

Ami Attali, Henry Ford- Detroit

Arvind Palanisamy, Washington University

Melissa Bauer, Michigan Medicine

Carlo Pancaro, Michigan Medicine

Dan Biggs, University of Oklahoma

Mohamed Tiouririne, University of Virginia

David Swastek, St. Joseph Mercy Ann Arbor

Brandon Togioka, Oregon Health Science University

Eric Davies, St. Joseph Mercy Oakland

Joshua Younger, Henry Ford, Detroit

Ghislaine Echevarria, NYU Langone

Marie-Louise Meng, Duke

Ronald George, University of California- San

Francisco

Ashraf Habib, Duke

Jenifer Henderson, St. Joseph Oakland

Tom Klumpner, Michigan Medicine

Rachel Kacmar, University of Colorado

Nirav Shah, MPOG Associate Director

Joanna Kountanis, Michigan Medicine

Kate Buehler, MPOG Clinical Program Manager

Carlos Delgado Upegui, University of

Washington

Meridith Bailey, MPOG QI Coordinator

Stephanie Lim, University of California- San

Francisco

Brooke Szymanski, MPOG QI Coordinator

1. OB Anesthesia Subcommittee Leadership

a. Thank you to Dr. Rachel Kacmar (University of Colorado) and Dr. Dan Biggs (University of

Oklahoma) for serving as leaders of the OB Anesthesia subcommittee!

2. 3/2020 Meeting Recap

a. ABX 01-OB: Antibiotic Timing for Cesarean Delivery Specification Review

i. Committee agreed on including emergent cases, but with a more forgiving

timeframe for passing

ii. Committee agreed on adjusting appropriate timeframe for azithromycin (60 min

before incision through anes end) and clindamycin (within 60 minutes before

surgical incision). All other antibiotics must be administered ‘within 60 minutes

before incision’ (except vanco, which is 120 minutes) to ‘pass’ the measure

b. Antibiotic selection measure

i. Determined to be lower priority, may be useful to develop later in collaboration

with surgical colleagues

c. Prolonged Hypotension Measure Specification Review

i. Discussed timeframe and applicable BP cutoffs for specific populations

ii. Committee agreed on lack of standardization, this measure will remain

informational only

iii. Committee agreed on separate measures for patients with pre-eclampsia and

those without

3. New OB Concepts Available for Mapping

a. Fetal heart rate- Decelerations (50238)

b. Fetal heart rate – Accelerations (50239)

c. Fetal heart rate category (3166)

4. Epic OB Variables – MPOG is working with Epic to include more variables in the MPOG extract

potentially used at your site.

a. *See slide 8 of presentation for list of variables

b. Contact the coordinating center if you would like to be connected with the individual

who is responsible mapping for your site

5. AKI Toolkit

a. MPOG is happy to announce the release of its 4

toolkit: Avoiding Kidney Injury. This

toolkit reviews the pathophysiology and definitions of kidney injury as well as

recommendations for perioperative care.

i. This toolkit includes a one-pager summary and slide presentations

ii. Overview, Pathophysiology & Definitions of Kidney injury

iii. Recommendations to prevent Kidney injury: Adult Surgical Patients

1. Pediatrics

2. Cardiac

3. Obstetrics

4. https://mpog.org/akitoolkit/

6. AKI-01

a. AKI 01 measures the percentage of cases where the baseline creatinine does not

increase more than 1.5 times within 7 postoperative days or the baseline creatinine

level does not increase by = 0.3 mg/dL within 48 hours postoperatively.

b. Currently includes pre-eclampsia patients who undergo cesarean delivery

i. AKI 01 excludes non-operative procedures including labor epidurals currently

ii. Cesarean deliveries with pre-eclampsia are included

c. References:

i. Van Hook JW: Acute kidney injury during pregnancy. Clin Obstet Gynecol 2014;

57:851–61

ii. Huang C, Chen S: Acute kidney injury during pregnancy and puerperium: a

retrospective study in a single center. BMC Nephrol 2017; 18:146

iii. Arulkumaran N, Lightstone L: Severe pre-eclampsia and hypertensive crises.

Best Pract Res Clin Obstet Gynaecol 2013; 27:877–84

d. Discussion: Should AKI 01 exclude pre-eclampsia patients?

i. Arvind Palanisamy (Washington University) Separate pre-eclampsia patients

with severe features from those without severe features to know if those

without severe features are at increased risk for AKI. Wash U obtains preop

chemistries on all OB patients.

ii. Dan Biggs (University of Oklahoma) The only patients with preoperative SCr labs

are those with pre-eclampsia. If you include, you will not be measuring normal

patients. Should include patients with pre-eclampsia for this measure but

understand, we are looking at a high-risk population.

iii. Joshua Younger (Henry Ford Detroit) Agrees with including pre-eclampsia

patients. Only going to have SCr labs for patients that are already of a concern.

From a research perspective, look at patients with pre-eclampsia with severe

features but those that don’t relate to renal function. i.e. neuro issue of liver

dysfunction.

iv. Ashraf Habib (Duke) Agree. SCr mostly done on patients with pre-eclampsia with

severe features. Interesting research question for AKI in pre-eclampsia patients.

1. Nirav Shah (Michigan Medicine) For AKI-01 we exclude patients where

there is no baseline creatinine available. Agree there is a lot of room for

research of AKI in pre-eclampsia patients. Contact the Coordinating

Center if interested in this.

v. Brandon Togioka (Oregon Health Science University) – Concerns comparing

performance across institutions; dependent on patient populations which are all

treated very differently - especially midwife patients which rarely have preop

labs drawn. Primarily would be including high-risk patients and wouldn’t

accurately reflect practice across the entire institution.

1. Nirav Shah (Michigan Medicine) We perform risk adjustment for our AKI

measure; not 100% accurate. Primary purpose of this measure is case

review at the institution level – less so to provide feedback to individual

providers. For benchmarking it is tough to make generalizations

especially at a subspecialty level. Excluding pre-eclampsia patients from

AKI-01 will not make a major difference in the overall performance

score. MPOG bias is to include them. If we decide to do research or QI,

it does make sense to be able to notice when those patients have AKI

whether or not we expect it.

vi. Angel Martino (Sparrow Health System) – Is it possible to look at pre-eclampsia

patients with severe features vs. pre-eclampsia patients without severe

features? There is a qualifier in Epic documentation for this at Sparrow.

1. Nirav Shah (Michigan Medicine) ICD 9/10 code could also be a

possibility

2. Arvind Palanisamy (Washington University) – It was mandated that

academic institutions document severe features starting in 2019. Most

sites should have this data. Criteria remains the same for pre-eclampsia

w/severe features and ‘severe pre-eclampsia’. Removed mild/moderate

pre-eclampsia from definition to prevent the interpretation that pre-

eclampsia was on a continuum. There now exists ‘pre-eclampsia’ which

is simply BP>140/90 without other symptoms and ‘pre-eclampsia with

severe features’ which is hypertension + additional symptoms

(headache, kidney injury, vision changes, impaired liver function, etc.)

vii. Joshua Younger (Henry Ford Detroit) We could also look at patients who receive

magnesium. Not guaranteed to have severe features

viii. Final conclusion: Exclude patients from AKI 01 who have pre-eclampsia with

severe features using ICD-10 codes (O14.1%)

7. ABX 01-OB: Antibiotic Timing for Cesarean Delivery Performance Review

a. First MPOG OB measure released 7/15/20!

i. Variation across institutions for this measure: Outlier low performing sites may

be due to documentation/data extract rather than reflective of actual practice.

Some sites do not consistently document procedure start for cesarean deliveries

or document their antibiotics as note rather than on the MAR. Both of these

scenarios result in flagged cases and lower performance on the dashboard. This

measure is specifically looking at antibiotic timing and includes the most

common antibiotics administered for cesarean delivery. See specification for

more details.

ii. Encourage all members of the committee to log into the ASPIRE dashboard to

examine performance and cases for their own institution. If access is needed,

contact Coordinating Center.

b. Dan Biggs (University of Oklahoma) How many people are giving Azithromycin for C-

sections?

i. Arvind Palanisamy (Washington University) – Azithromycin paper came from

Wash U so this institution may be an outlier as compliance is 100%

ii. Clarification from Coordinating Center: this measure does not examine antibiotic

selection, only timing.

c. Joshua Younger (Henry Ford Detroit) Azithromycin always seems like a secondary

thought after Cefazolin. Azithromycin is usually always given but not always before

incision.

i. Arvind Palanisamy (Washington University) – For emergent procedures, patients

should get azithromycin. For antepartum C-sections level 2and level 3, these

patients don’t get azithromycin prior to incision at WashU. Most cases still meet

criteria to receive azithromycin but may not be administered prior to incision.

For the clinical trial that was completed it was done in a very controlled fashion

but may not be applicable to all clinical scenarios.

8. BP-04-OB: Prolonged Hypotension Measure Specification Review

a. Based on prelim data from 7/1/19-4/30-20

i. 37,739 Total C-sections (scheduled or conversion from vaginal delivery)

ii. 3,800 (10%) of the total c-sections were identified as having pre-eclampsia using

our phenotype

iii. 914 (24%) have a baseline BP in MPOG.

b. Measure Time Frame  Total cumulative minutes of hypotension will be resulted for

two time periods: spinal placement to delivery and delivery through anesthesia

end. Need to determine the threshold for number of minutes of hypotension that

results in a flagged case.

c. Discussion:

i. Joshua Younger (Henry Ford Detroit) Proposed two time frames: pre-delivery and

post-delivery and women shouldn’t have sustained hypotension for more than

10 minutes if patient has baby in utero.

ii. Nirav Shah (Michigan Medicine) QI dashboard visualization for this measure

would display blood pressures for two, possibly three time periods:

1. Spinal placement to delivery and

2. Delivery to anesthesia End.

3. Could also display total cumulative minutes across the entire case

iii. Joshua Younger (Henry Ford Detroit) – I would suggest displaying all 3 phases:

pre-delivery, post-delivery and cumulative minutes of hypotension for the full

case: spinal placement to anes end.

iv. Rachel Kacmar (University of Colorado) – I agree it is best to split the data out.

Most crucial time point to maintain normal blood pressure is spinal to delivery of

fetus. A little more leeway on 10-20% blood pressure drift post-delivery as the

fetus is no longer at risk.

v. Ronald George (UCSF) – Spinal placement to delivery of baby should be the focus

vi. Arvind Palanisamy (Washington University) – Agree- spinal placement to

delivery time period is most critical. Post-delivery may not be a useful marker.

Will see dips after delivery due to discontinuing infusions. We should focus on

the pre-delivery phase.

vii. Mohamed Tiouririne (University of Virginia) – I propose we decide between a

direct measure or indirect measure. Standard guidelines are to initiate infusions

of phenylephrine or norepinephrine. Can we use that marker instead rather than

actual blood pressure? Would be more of a global marker.

1. Ronald George (UCSF) – Very interesting point. We do have good

evidence to support the use of these medications to manage

hypotension.

2. Ashraf Habib (Duke) – We should look at 1) process of if you are using

medications and 2) did you achieve the target or not. If we can look at

both the process (medication administration) and outcome

(hypotension)- would be best rather than one or the other. Sustained

hypotension for 10 minutes is way too long while fetus in utero.

a. Arvind Palanisamy (Washington University) – Not a lot of data

to know what range of hypotension could impair perfusion of

fetus. 3 minutes seems to be the point where there would be a

consequence to the fetus. Will share some of the literature

which uses 3 minutes.

3. Nirav Shah (Michigan Medicine) – Using 3 minutes would mean a single

measurement of systolic < 90mmhg would be consequential. Would

need to cycle the BP cuff sooner. We extract minute by minute BP values

for arterial lines and we extract non-invasive BP values with their time

stamps (3 min, 5 min etc). We do further processing at the coordinating

center via artifact reduction algorithm and carry forward a blood

pressure until the next one becomes available. For details, visit MPOG

Blood Pressure Observations algorithm:

https://collations.mpogresearch.org/Detail.aspx?name=Blood%20Press

ure%20Observations

a. Multiple blood pressures: Instances where there are two blood

pressure monitoring methods, the higher MAP will be used to

determine measure compliance.

b. Artifact: Artifact readings will be identified and removed from

final measurement calculation. Artifact processing: if systolic

and diastolic blood pressures are present, the values must be at

least 5 mmHg apart; otherwise the values will be excluded. MAP

values less than 10 are excluded.

i. Each incidence of hypotension will count for a max of 5

minutes if there is a gap in blood pressure

measurement

ii. Average cumulative minutes of hypotension from spinal

placement to delivery of the last neonate (if more than

one) will be resulted as one number for the institution.

Individual cases will show the total cumulative minutes

of hypotension for this time period.

iii. Average cumulative minutes of hypotension from

delivery to anesthesia end will be resulted as a second

number for the institution. Individual cases will show

the total cumulative minutes of hypotension for this

time period.

4. Ashraf Habib (Duke) – frequency of BP measurements is a good measure

as well.

d. Measure Description Discussion:

i. Ashraf Habib (Duke) - Cases with SBP<90 for greater than 3-5 minutes would

make the most sense for flagging cases

ii. Arvind Palanisamy (Washington University) – 5 minutes should be ok instead of

3 minutes.

iii. Angel Martino (Sparrow Health System) – Is it possible to note inside the

measure whether or not treatment was instituted for the low blood pressures

sustained? Then you are capturing process and outcome data.

1. Nirav Shah (Michigan Medicine) – yes we can provide this information in

the case details tab of the dashboard. For flagged cases, we can indicate

whether a vasopressor was initiated.

iv. Joshua Younger (Henry Ford Detroit) What about patients that are 90/50 at

baseline? How do we account for that?

1. Nirav Shah (Michigan Medicine) Can add baseline as information in case

details. However, pre-anesthesia blood pressures aren’t a great option

as relative decreases will be difficult to determine.

v. Joshua Younger (Henry Ford Detroit) – Could use one BP prior to Anesthetic in

the OR

1. Ashraf Habib (Duke): Typically, the blood pressures before anesthesia

are elevated and not reflective of the patient’s baseline

2. Angel Martino (Sparrow): Agree

9. Meeting Conclusion

a. Meeting minutes and outstanding agenda items will be posted to the forum for further

discussion:

i. Given the low # of pts identified to have preeclampsia and the low % with

baseline BP, should preeclampsia patients be excluded from this measure? Or

included but noted to have pre-eclampsia?

ii. Should scheduled and conversion cesarean delivery cases be treated the same

for BP 04?

iii. Should patients with HELLP syndrome be excluded or treated as pre-eclampsia

patients?

Meeting adjourned at 1402