Comparison of Umbilical Venous and Intraosseous Access During Simulated Neonatal Resuscitation

1
Comparison of Umbilical Venous and Intraosseous
Access During Simulated Neonatal Resuscitation

• Anand Rajani, M.D.
• Perinatal Medical Group, Inc.
• Fresno, California
• Previous affiliation
• Fellow in Neonatal-Perinatal Medicine
• Stanford University School of Medicine
• Lucile Packard Childrens Hospital
• Palo Alto, California

2
Disclosure

• I have nothing to disclose.
• This work was supported by the Young Investigator
  Award from the Neonatal Resuscitation Program.

3
Background

• While 10 of newborns require some assistance to
  begin breathing, only 1 require extensive
  resuscitative efforts
• Less than 2 in 1000 births require administration
  of intravenous epinephrine1
• Proficiency in rapid umbilical venous catheter
  (UVC) placement is difficult to maintain

1. Perlman JM, Risser R. Cardiopulmonary
resuscitation in the delivery room. Arch Pediatr
Adolesc Med 199514920 5
4
Background

• Establishing umbilical venous access is
  frequently difficult
• Catheter setup
• Thoracic compressions
• Moving sterile field
• Data indicate that intraosseous needle (IO)
  placement is a safe and effective alternative
• Access times of 30-60 seconds in the pediatric
  setting2
• Pharmacokinetic data on IO epinephrine in newborn
  lambs suggest equal efficacy3

2. Zaritsky AL, Nadkarni UM, Hickey RW, et al.
PALS provider manual. Dallas (TX)7 American Heart
Association/American Academy of Pediatrics
2002 3. Ellemunter H, Simma B, Trawoger R, et al.
Intraosseous lines in preterm and full term
neonates. Arch Dis Child Fetal Neonatal Ed
199080F74-5.
5
Simulation

• Allows for the re-creation of high-risk, low
  frequency events in numbers that are useful for
  statistical analysis
• Can be video-recorded for further analysis
• No harm to real patients

6
Hypotheses

• Primary Null Hypothesis
• Ho IO and UVC placement will be established in
  equal time
• Secondary Null Hypothesis
• Ho IO and UVC placement will be established with
  equal rates of error
• Observational Null Hypothesis
• Ho Perceived ease of use will be equal for UVC
  and IO

7
Methods

• Recruited 40 healthcare practitioners of varying
  training levels from Lucile Packard Childrens
  Hospital at Stanford

Training Level N ()
Resident in Pediatrics 16 (40)
Fellow in Neonatology 6 (15)
Neonatal Hospitalist 5 (12)
Neonatal Nurse Practitioner 5 (12)
Attending Neonatologist 8 (20)
8
Methods

• Two standardized, videotaped simulated
  resuscitation scenarios in which intravascular
  access was indicated
• A nurse and RT confederate performed CPR while
  the participant established access
• Indistinguishable kits containing UVCs or IOs
  were available at the bedside
• Simulation was stopped once access established

9
Methods Study Design

• Prospective, blinded, randomized, 2x2 crossover
  design
• Randomized participants in separate blocks, by
  training level to perform either
• UVC/IO or IO/UVC
• Prior to the simulations, participants watched a
  video reviewing the necessary steps involved in
  placement of a UVC and IO needle

10
Methods Data Collection

• Using video recordings
• Placement Time
• Errors during placement
• 4 error categories were used for each modality
• Site preparation
• Device Preparation
• Location and depth
• Confirmation of access

11
Methods Data Collection

• Using questionnaire
• Users perception of technical difficulty (Likert
  scale from 0-10)
• Preference for IO or UVC, if any
• asked for reasons behind preference
• space left for additional comments

12
Analyses for Primary Hypothesis

• Ho IO and UVC will be established in equal time
• Test 1 t-test to evaluate for period effect
• Evaluate the difference in the two time periods
  of UVC/IO and IO/UVC
• There was no significant difference in placement
  times for UVC or IO relative to placement order

13
Analyses for Primary Hypothesis

• Test 2 Matched pairs t-test to evaluate for any
  difference in placement time between UVC and IO
• For placement time, IO was significantly faster
  (plt0.0001)
• Using ANOVAs, resident group was significantly
  faster than all other groups

14
UVC and IO placement by subgroup
Training Level (N) UVC Time (sec) IO Time (sec) p value
All subjects (40) 105 59 lt0.0001
Residents (16) 105 17 lt0.0001
Fellows (6) 86 73 0.4431
Hospitalists (5) 104 86 0.4195
NNPs (5) 120 92 0.1238
Attendings (8) 111 94 lt0.0326
15
Analyses for Secondary Hypotheses

• Ho IO and UVC will be established with equal
  rates of error
• No significant difference was found
• 3 errors in the IO group (site prep)
• 1 error in the UVC group (site prep)

16
Analysis of Observational Hypothesis

• Ho Perceived ease of use will be similar for UVC
  and IO
• UVC and IO found to be equivalent
• Residents (n16) found IO to be easier to place
  than UVC (p0.003)
• 25 (4) residents preferred IO 2 had no
  preference
• 22 participants preferred the UVC -- all cited
  familiarity as a reason for this preference
• difference in experience years vs. minutes!

17
UVC and IO perceived ease of use by subgroup
Training Level (N) UVC difficulty IO difficulty p value
All subjects (40) 4.6 4.3 0.6762
Residents (16) 6.5 4.75 0.0026
Fellows (6) 4.3 3.8 0.6462
Hospitalists (5) 4.4 6 0.2420
NNPs (5) 2.2 4.6 0.1856
Attendings (8) 1.8 2.5 0.1395
18
Discussion

• Difference between mean IO and UVC placement was
  0.76 minutes (46 seconds)
• Identifies differences in time to placement --
  does not account for how components are packaged
• Implications for NRP / Possible practice changes
• perhaps IO should also be taught and recommended
  as a placement technique (not shown to be
  inferior)
• UVCs could be recommended for use in tertiary
  care centers where there is consistent
  experience IOs may be more appropriate elsewhere

19
Conclusions

• For the primary hypothesis must reject Ho
• IO is faster than UVC
• For the secondary hypothesis must accept Ho
• no difference in rates of error
• For the observational hypothesis must accept Ho
• no difference in perceived ease of use

20
References

1. Perlman JM, Risser R. Cardiopulmonary
   resuscitation in the delivery room. Arch Pediatr
   Adolesc Med 199514920-5.
2. Zaritsky AL, Nadkarni UM, Hickey RW, et al. PALS
   provider manual. Dallas (TX)7 American Heart
   Association/American Academy of Pediatrics 2002
3. Ellemunter H, Simma B, Trawoger R, et al.
   Intraosseous lines in preterm and full term
   neonates. Arch Dis Child Fetal Neonatal Ed
   199080F74-5.
4. Sapien R, Stein H, Padbury JF, Thio S, Hodge D.
   Intraosseous versus intravenous epinephrine
   infusions in lambs Pharmacokinetics and
   pharmacodynamics. Ped Emerg Care 19928179-183.