0:09

Welcome to The Skeptics Guide

0:11

to Emergency Medicine. Meet them, greet

0:13

them, treat them, and street them.

0:16

Today's date is February 7th 2025

0:18

and I'm your skeptical

0:20

host, Dennis Wren. The title of

0:22

today's episode is, here we go up,

0:24

up, up, up, or lateral for infant

0:27

lumbar punctures. And our guest

0:29

skeptic is Dr. Lauren

0:31

Rosenfeld who is a

0:33

PGY3 emergency medicine resident

0:35

at George Washington University.

0:37

She is also a new podcast

0:40

host for Emergency Medicine Residents

0:42

Association or Amra Cast series.

0:45

Lauren, welcome to the S gem.

0:47

Thank you very much for having me.

0:49

I'm so excited to be here. And

0:51

Lauren, we actually met while we

0:53

were on shift together in the

0:56

emergency department at Children's and we

0:58

just bonded over our love of

1:01

podcast. So it's awesome to have you

1:03

here. Yeah, I'm super excited to be

1:05

here. Podcasting is a new passion of

1:07

mine. It is always fun to record,

1:09

always fun to meet new people and

1:12

do some learning. Well now that you

1:14

have entered this world, I hope you meet

1:16

many, many more friendly podcasters. I

1:18

think they're just such a nice

1:21

group of people to get to

1:23

know and share tips and tricks

1:25

with. Absolutely. All right, but this is

1:27

a critical appraisal medical podcast.

1:29

So let's get down to it.

1:31

We are talking about lumbar punctures

1:34

today and I understand you brought

1:36

us a case. I have a five-day-old

1:38

girl who's brought into the

1:41

emergency department for a fever.

1:43

She's brought in by her parents.

1:45

She was full term, seemed to

1:47

be doing relatively well after the

1:49

family brought her home. Mom had an

1:52

uneventful pregnancy delivery, but

1:54

today, her parents thought she

1:56

was feeling a little warm. She took

1:58

her temperature and... it

2:00

was 101, an actual fever

2:02

above 38 Celsius. So they

2:05

called their pediatrician and

2:07

guess what the pediatrician

2:09

told them. Bring them on in.

2:12

She did warn them that their

2:14

baby might need a lumbar

2:16

puncture. And the worried father sits

2:18

there asking me, what is a

2:21

lumbar puncture? Will it hurt? Mom

2:23

asked, is it like when I

2:25

got an epidural before

2:27

delivery? Will you set her out for

2:30

it? She can't even sit yet. She's

2:32

five days old. And we've covered

2:34

the topic of febral infants

2:36

lumbar punctures before on the

2:38

S-JIM. However, we typically focused

2:41

on the febral infant part

2:43

and today we're talking more

2:45

about the performance of the

2:47

procedure of a lumbar puncture

2:49

on babies. In the ED,

2:52

lumbar punctures are typically

2:54

performed in infants with

2:56

a fever and the

2:59

evaluation for invasive bacterial

3:01

infections, including meningitis. Now

3:03

Lauren, in your experience, is performing

3:05

an LP on a baby the

3:08

same as performing one on

3:10

an adult? Are adults big babies?

3:12

Yes, pretty much. But every LP

3:15

is a little bit different, whether

3:17

it's in a child or an

3:19

adult. I think babies are actually,

3:21

and we're talking, the five day-old

3:24

babies, are actually a little bit

3:26

more predictable because you have

3:28

a little bit more control

3:31

over the situation. Adults are

3:33

larger, stronger, usually, and often

3:35

require a little bit more

3:38

reassurance than children throughout the

3:40

procedure. Now I still remember one of

3:42

the first procedures I ever learned how

3:44

to do and actually performed was a

3:47

lumbar puncture on a baby that was

3:49

just a couple days old and there

3:51

are always many thoughts bits of advice

3:53

depending on who you're talking to and

3:55

who's teaching you about what's the proper

3:58

position when should you remove the dialect

4:00

from the spinal needle. What kind

4:02

of annual geezia should you use?

4:05

All that kind of stuff. Yes,

4:07

there are so many options and

4:09

multiple positions to set up the

4:11

spinal tap. Commonly, patients can

4:13

be placed on their side in

4:15

the left lateral decubitous or right

4:18

lateral decubitous. Bend the neck so

4:20

the chin is close to the

4:22

chest. Hunch the back and bring

4:24

the knees toward the chest to

4:26

approximate the fetal position. Alternatively,

4:29

patients may also sit upright

4:31

and bend their head and shoulders

4:34

forward. I recently did this with

4:36

an adult and my attending said

4:38

it was like a hunched over

4:40

cat. Oh, interesting. Well, I guess

4:42

the good thing is for these

4:44

super young babies, the fetal

4:46

position is just kind of

4:49

self-explanatory. It's close to what they tend

4:51

to do or what they just did. But

4:53

when it comes to kids, most of the

4:55

time, you know, we are also still relying

4:57

on someone else to help hold that

4:59

baby in these positions as we're

5:01

performing the LP. And sometimes, depending

5:04

on how long it takes you

5:06

and how much force they're using,

5:08

I've seen episodes where they

5:10

have oxygen sea saturations because

5:12

they just get held for that long.

5:14

Lauren, do you prefer any position

5:16

over the other? You know. I think the upright

5:18

seated position is nice, so there

5:21

is less distortion of the spinal

5:23

anatomy, and you can kind of

5:25

feel on both sides when you're

5:27

trying to get that perfect position.

5:29

It also allows for easier withdrawal

5:31

of the fluid, whereas lying on

5:34

the side results in less reliable

5:36

anatomical landmarks, but I will

5:38

say those opening pressures are

5:41

notoriously unreliable when measured in

5:43

the seated position. Oh yes,

5:45

opening pressures and we're not

5:47

even talking about those today.

5:49

But that being said, I've seen

5:51

so many people do LPs in

5:54

all kinds of different positions. Sometimes,

5:56

honestly, I think you can have

5:58

almost everything right. but still

6:00

just miss and be unsuccessful.

6:03

Yeah, so that brings us

6:05

to our clinical question. How

6:07

does positioning of infants during

6:10

a lumbar puncture lateral decubitous

6:12

versus sitting versus prone affect

6:14

success rates and adverse events?

6:17

And what's our reference? Passano

6:19

S at all. positioning for

6:21

lumbar puncture in newborn infants.

6:23

It was from a Cochran

6:26

database systemic review in December

6:28

2023. Let's talk through our

6:30

Peacot questions. What was the

6:33

population they looked at? Population

6:35

they looked at was pre-term

6:37

and term infants of post

6:40

menstrual age up to 46

6:42

weeks and zero days. Age

6:44

4.9 hours to five weeks

6:47

old. What was their intervention?

6:49

So they looked at infants

6:51

positioned in a lateral decubitous

6:53

position and the comparison infants

6:56

positioned in a sitting position

6:58

or prone position Now we

7:00

will note here that they

7:03

actually couldn't find any studies

7:05

comparing sitting position to prone,

7:07

but let's talk about their

7:10

outcomes now. What was the

7:12

primary outcome they were looking

7:14

at? So of course primary

7:17

outcome is success So successful

7:19

lumbar puncture on the first

7:21

attempt with less than 500

7:23

red blood cells, so a

7:26

clean tap, total number of

7:28

lumbar puncture attempts, successful or

7:30

unsuccessful. They also looked at

7:33

episodes of braticardia defined as

7:35

a decrease in heart rate

7:37

of more than 30% below

7:40

baseline or less than 100

7:42

beats per minute for 10

7:44

seconds or longer. I think

7:47

that was three primary outcomes.

7:50

There can be only one.

7:52

Okay, let's talk about their

7:55

secondary. What were the secondary

7:57

outcomes? So we had time

7:59

to perform LP episode. of

8:01

desaturation so that would be

8:03

an SPO2 of less than

8:05

80%. Apnea, need for pain

8:07

or sedation medication, skin changes

8:09

at the LP site, infection

8:12

rate related to the LP,

8:14

pain, and parental satisfaction. And

8:16

finally what type of study

8:18

was this? Of course it

8:20

was your systematic review meta

8:22

analysis. All right and the

8:24

authors concluded. Oh boy, this

8:26

was a really long conclusion.

8:29

When compared to sitting position,

8:31

lateral to cubitous position probably

8:33

results in little to no

8:35

difference in successful lumbar puncture

8:37

procedure at first attempt. None

8:39

of the included studies reported

8:41

the total number of lumbar

8:43

puncture attempts as specified in

8:46

this review. Furthermore, infants in

8:48

a sitting position likely experience

8:50

less episodes of radicardia. and

8:52

oxygen desaturation, then in the

8:54

lateral decubitous. And there may

8:56

be little to no difference

8:58

in episodes of apnea. Lateral

9:00

decubitous position results in little

9:03

to no difference in time

9:05

to perform the lumbar puncture

9:07

compared to sitting position. Pain

9:09

intensity during and after the

9:11

procedure was reported using a

9:13

pain scale that was not

9:15

included in our pre-specified tools

9:17

for pain assessment due to

9:20

its high risk of bias.

9:22

Most study participants were term

9:24

newborns, thereby limiting the applicability

9:26

of these results to preterm

9:28

babies. When compared to prone

9:30

position, lateral decubitous position may

9:32

reduce successful lumbar puncture procedure

9:34

at first attempt. Only one

9:37

study reported on this comparison

9:39

and did not evaluate adverse

9:41

effects. Further research exploring harms

9:43

and benefits and the effect

9:45

on patients pain experience of

9:47

different positions during lumbar puncture

9:49

using validated pain scoring tools

9:51

may increase the level of

9:54

confidence in our conclusions. I

9:56

Guess that's what you get

9:58

when you have so many

10:00

outcomes that you're looking at.

10:02

But moving on to

10:05

our quality checklist, first

10:07

question, was the clinical

10:09

question sensible

10:12

and answerable? I would

10:14

say yes. Was the search

10:16

for studies detailed and

10:18

exhaustive? As you heard, yes.

10:21

Were the primary studies

10:23

of high methodological

10:25

quality. So we covered

10:27

this a little bit. I would have

10:29

to say no. And we'll talk even

10:32

more about this later on. Was the

10:34

assessment of the

10:36

studies reproducible? Yes. Were the outcomes

10:39

clinically relevant? Yes.

10:41

Use it in my practice on

10:43

the regular. Was there low

10:46

statistical heterogeneity for the

10:48

primary outcomes? I would

10:50

say no. The successful

10:52

LP at first attempt

10:54

has a moderate statistical

10:56

heterogeneity. For many other

10:59

outcomes assessing for heterogeneity

11:01

was not applicable. Was the

11:03

treatment effect large enough

11:05

and precise enough to

11:07

be clinically significant? Unfortunately

11:10

not. And were there any

11:12

financial conflicts of interest?

11:15

No, not for us residents

11:17

especially. All right, fair enough. Nobody

11:19

got paid by big LP to

11:21

do this study. Moving on

11:23

to their results, so they

11:26

included five studies that had

11:28

close to 1,500 participants, the

11:30

mean gestational age of the

11:32

infants, and the included studies

11:35

ranged from 31 to 41

11:37

weeks, with the largest study enrolling

11:39

mostly term newborns. The mean

11:42

postnatal age at the time

11:44

of procedure completion ranged from

11:46

4.9 hours to 5 weeks. Lauren, what

11:48

was the key result? There was not

11:50

that much difference in LP success

11:53

with the lateral decubitous position compared

11:55

to the other positions. However, I

11:57

will say that the lateral decubitous...

12:00

disposition may be associated

12:02

with more episodes of

12:04

braticardia and desaturations. Let's break

12:06

this down now based on the

12:08

outcomes. So looking at their primary

12:10

outcomes, like you mentioned Lauren, there

12:12

was no difference in LP success

12:14

between lateral decubitous versus sitting position

12:17

and we'll have the numbers in

12:19

our show notes. And as we

12:21

just said that lateral decubitous

12:23

positioning did increase those episodes

12:25

of braticardia and the number needed

12:27

to harm was 33. The lateral

12:30

decubitous positioning also

12:32

increased episodes of desaturation

12:35

and that had a

12:37

number needed to harm of

12:40

17. Looking at their

12:42

secondary outcomes those were

12:45

again key patient-centered outcomes

12:47

like pain, infection, sedation

12:50

needs, and parental satisfaction.

12:52

A lot of those

12:54

were actually not reported.

12:57

All right, Lauren, are you ready for

13:00

my favorite section? I'm ready. Let's

13:02

talk nerdy. First nerdy point goes

13:04

to you. Let's talk about these

13:06

studies they actually included in

13:08

this. Overall, they only found

13:10

five studies to include in

13:13

this review. Four were randomized

13:15

control trials. And one was

13:17

a quasi randomized control trial.

13:19

Most of the data for this

13:22

review came from one study

13:24

that had a thousand eighty

13:26

two participants. which is around

13:28

73% of all the participants

13:30

included in the review. When we

13:32

look at the outcomes, they were trying

13:34

to assess. Most of the time, only

13:36

two or three studies reported the outcome

13:39

of interest, making testing

13:41

for heterogeneity, as we

13:43

mentioned, challenging. Our second

13:45

nerdy point is about the

13:47

certainty of the evidence. So

13:50

even though they included five

13:52

studies, these studies only included

13:54

a total of 1,476 patients.

13:57

So because of that limited

13:59

data, Many of the outcomes

14:01

they were looking at were moderate

14:03

or low certainty of evidence. And

14:05

there was only one outcome that

14:07

achieved high certainty, which was that

14:09

there was little to no difference

14:11

in time to perform lumbar puncture

14:13

when comparing lateral to cubitus to

14:15

sitting position. The time to perform a

14:17

lumbar puncture may vary quite a

14:19

bit depending on the experience of

14:22

the clinician performing the procedure.

14:24

It may be the case that the

14:26

longer it takes to perform the LP,

14:28

the more risk of those adverse events

14:31

like desaturations or predicardia occurs because the

14:33

baby is all screnched up in that

14:35

position. Our third nerdy

14:37

point is about patient-oriented

14:40

versus monitor-oriented outcomes. So their

14:42

outcomes of interest were a

14:44

mix of patient-oriented outcomes or

14:46

poos, which we love, and

14:49

monitor-oriented outcomes moves. Now one

14:51

fairly important patient-oriented outcome that

14:53

wasn't reported across the

14:55

studies included the number

14:57

of lumbar puncture attempts. Now

15:00

I would say as a parent and a

15:02

caregiver, this is important. And if

15:04

you are on the receiving end of this

15:06

lumbar puncture, you would probably want

15:08

to know that too. because I

15:10

don't know how happy I would

15:12

be if somebody was successful with

15:14

their LP, but in the process they

15:17

turn my kid into a pin cushion.

15:19

I typically will stop making attempts

15:21

if I can't get that CSF

15:23

after three tries. Absolutely. And the

15:25

outcomes of desaturations and

15:27

braticardia are monitor-oriented outcomes.

15:30

The definition for what

15:32

counted as a desaturation

15:34

or braticardic episode varied

15:36

or was not reported across the

15:38

studies included. The authors define

15:41

desaturation as a pole socks

15:43

less than 80% which is

15:45

pretty generous because once it

15:47

hits, you know, below 90

15:49

I start getting a little

15:51

nervous. You and me both.

15:53

With no minimum duration or

15:55

apnea as interruption and breathing

15:57

for more than 20 seconds.

16:00

that's, I guess, a bit more standard.

16:02

It is also unclear if these

16:04

desaturations or braticardic episodes

16:06

were sustained self-resolving or

16:08

even required intervention. Is

16:10

it accurate to attribute

16:12

these events to the

16:14

LP procedure itself even? Yeah,

16:16

and that's a nice segue into

16:18

nerdy point number four, which is

16:20

about the indications for performing that

16:22

LP in the first place. There

16:25

was variation in the populations that

16:27

were being studied in each of

16:29

the included studies. One study

16:31

only included sick neonates. One

16:33

study included infants one to

16:35

90 days undergoing LP in the

16:37

emergency department. It did not

16:39

specify the indications. One study

16:42

included free term infants who

16:44

received LP for spinal anesthesia

16:46

before inguinal hernia repair. The largest

16:48

study included infants 27 to

16:50

44 weeks corrected gestational age.

16:52

and most of these study

16:54

participants were included due to

16:57

concerns for infection or sepsis.

16:59

Now the difference in the included

17:01

populations could have also impacted

17:03

the results. For example, it's

17:05

possible that the sicker babies may

17:07

be at more risk of having

17:09

episodes of desaturation, bratocardia, apnea, compared

17:11

to those who were either well-appearing

17:13

febrel infants or they were just

17:15

receiving that LP for anesthesia. Our

17:18

fifth and last point is

17:20

about unmeasured or unreported or

17:22

unreported... confounders. Yeah, we spoke earlier

17:25

in the show about how there

17:27

were a lot of factors to

17:29

consider when performing a lumbar puncture.

17:31

That can include adequate

17:33

analgesia, technique of the person

17:35

holding the baby, early

17:37

stylet removal, and experience

17:39

of the performing physician.

17:42

Yeah, I know I missed my

17:44

first couple of attempts at the

17:46

lumbar puncture and I first started.

17:48

Yeah, you're telling me. The

17:50

positioning of the baby is just

17:52

one of the factors that comes

17:54

into play in determining the success

17:56

of a procedure. I think the authors

17:58

tried to include... of these factors in

18:01

the review, but many simply weren't

18:03

reported in the original studies.

18:05

All right, and I think we

18:07

should end this with five other

18:09

potential biases, since we love the

18:12

number five here. So I'll start

18:14

off about just overall. There was

18:16

possibility of selection bias. If these

18:18

patients were not randomly assigned to

18:20

the different positions, the baseline characteristics

18:23

could differ affecting the outcomes. And

18:25

of course, performance bias,

18:27

operators may have had

18:30

varied levels of experiences

18:32

with different positioning techniques

18:34

influencing success rates. Third

18:36

bias is detection bias.

18:38

The assessment of success

18:40

and complications such as

18:42

bradicardia desaturation may not

18:44

have been blinded leading

18:46

to potential measurement bias.

18:49

Number four. Heterogeneity in

18:51

technique. Variations in how LPs

18:53

were performed. The needle type

18:55

use of ultrasound guidance

18:58

may have introduced inconsistencies

19:00

in the pulled results.

19:03

And finally, publication

19:05

bias. Studies with negative

19:07

or non-significant findings, God

19:09

forbid your P value

19:12

is over 0.05, may have

19:14

been underrepresented, skewing

19:17

those conclusions. So

19:19

Lauren with all that being said

19:22

can you comment on the author's

19:24

conclusion compared to the SGM

19:26

conclusion? Well in summary we agree

19:28

with those authors conclusions. And

19:30

give us the SGM bottom line. There is

19:33

no evidence that any one

19:35

position increases the success rate

19:37

of LP and infants. Stick

19:39

with what you're most comfortable

19:42

with I guess. And that five-day-old

19:44

that you had presented at

19:46

the beginning of the episode,

19:48

how, what's going on? What

19:50

are you doing for that, baby?

19:52

So yeah, let's come back

19:54

to our five-day-old. So after

19:56

obtaining blood and urine studies,

19:58

you explain the... process of a

20:01

lumbar puncture to the family and

20:03

the reason behind why you're even

20:05

performing it. The parents consent to

20:08

the procedure and you follow all

20:10

the appropriate sterile procedures and have

20:12

someone even help you hold the

20:14

baby in the position that you're

20:17

most accustomed to performing

20:19

an LP. So for me that

20:21

was the sitting position and so

20:23

after I successfully obtained the CSF

20:26

and started Empiric antibiotics, we admitted

20:28

that baby to our hospital. Fantastic.

20:31

Now when it comes to the clinical application

20:33

of this study, the next time we're

20:35

working together on an ED shift and

20:37

we have to do a lumbar puncture

20:40

on a small baby, what are you doing?

20:42

You're sticking with that sitting position?

20:45

You know what? Let's just flip a coin.

20:47

Because to be honest, I've been

20:49

doing most of them in either

20:51

sitting position and more recently in

20:53

the left lateral decubitous. So whatever

20:55

I'm doing I should probably

20:58

practice the alternative. Yeah be versatile in

21:00

your technique and I feel like this probably

21:02

goes to say for all of the procedures

21:04

that we do in the ED right like

21:07

you try it in the way that you

21:09

are most accustomed to and most comfortable with

21:11

and if you don't get it which sometimes

21:13

you won't take a moment troubleshoot

21:16

and figure out what's the thing that

21:18

I'm going to change in this next

21:20

attempt that will optimize my success but

21:23

definitely don't just insist and do the

21:25

same thing over and over again.

21:27

Especially when you have an expert

21:29

just right there over your shoulder. Well,

21:32

even experts miss it sometimes too.

21:34

We are all imperfect. Lauren, how

21:36

do you explain this to the family?

21:39

What do you tell them? Well, what

21:41

I'll be telling this family and

21:43

families in the future is I'm

21:45

so sorry you are in the

21:47

emergency department so shortly after getting

21:49

home with your brand new baby.

21:51

At this age, fevers and babies

21:53

worry us because she's so young. We

21:56

don't know if she's having a fever because

21:58

she had a virus or... she has

22:00

bacteria in her urine or her

22:02

blood. The last thing we test is

22:05

the CSF fluid or cerebral

22:07

spinal fluid. This is the

22:09

fluid that surrounds her brain and

22:11

we do not want to miss

22:13

an infection there. Let's talk a

22:15

bit more about what the process

22:18

of a lumbar puncture is to

22:20

get that CSF and what that'll

22:22

look like. Thanks

22:24

so much for all that

22:26

nerdy discussion Lauren and I

22:29

hope you enjoy your experience

22:31

podcasting with us on the S

22:33

gem Always I can't wait to

22:36

hear it go live and to

22:38

hear even more from a skeptics

22:40

guide to emergency medicine And before

22:43

we go do you mind giving us

22:45

the S gem tagline? Remember

22:48

to be skeptical of anything

22:50

you learn, even if you

22:52

heard it on the Skeptics

22:54

Guide to Emergency Medicine.

22:56

Talk to everyone next

22:58

time.