Study #1 (Zengin S, et al. ):  This was a prospective study of 137 patients in cardiac arrest on arrival or during their encounter in one emergency department in Turkey from July 2016 to Sept 2017.  Patients were eligible if they were >16 years old.  Using a specially designed curtain system that descended from the ceiling, three distinct methods were used to obtain pulse checks simultaneously during cardiac arrest at three different time intervals (initial, 15 min, post-code) during the arrest:  cardiac ultrasonography, doppler ultrasonography, and manual pulse palpation method.  The pulse checks were performed by doctors of unknown training level or specialty.  The same two doctors performed ultrasound in every cardiac arrest and had a total of 24 hours of theoretical and applied focused ultrasound training prior to the study.  The objective of the study was to compare the efficiency between these three methods in terms of accuracy, speed, and effectiveness.  Time was in calculated in seconds and accuracy was calculated by percentage of false negative and false positive rates of detecting a pulse with manual and doppler checks as compared to ultrasound cardiac activity.  Findings included statistically significant difference in duration and accuracy (p < 0.001) between the three methods with cardiac ultrasonography being the fastest and most accurate, followed by doppler ultrasonography, followed by manual pulse palpation.

Strengths: This was a highly creative design in terms of the physical layout of the researchers behind curtains but there is much information missing in the design and methods description. 

Limitations: Were all three stations blinded to the cardiac monitor?  Were they wearing earplugs so they could not hear each other’s responses and/or the actual code team leader?  And if not, there could absolutely be bias involved, especially on the palpation method.  The terms were also not well defined, for instance, true versus false PEA were nearly clearly defined.  A pulse that could not be detected in 10s was logged as 11s to perform statistical calculation…did this significantly influence the results?  And in that same light, how were the pulse checks exactly timed?  Was there another person standing by each one of the three residents with a stopwatch?  Did they time themselves?  If so, this could also introduce another bias.  The also did not identify appropriate number of patients for 80% power (or it was at least not discussed).

Areas of improvement:  Be more specific in the methods section to decrease potential bias while three doctors are simultaneously performing pulse checks using three different mediums but only separated by curtains.  Also establish clear methods on how each of the three modes of pulse checks are timed as well as how the calculations were computed.

Study #2 (Badra K, et al.):  This was an IRB-approved prospective randomized controlled crossover non-inferiority trail of 111 participants that examined the speed and confidence of identifying carotid pulses in live subjects using manual versus ultrasound methods from July 2017 to January 2018 in an academic tertiary care hospital in Canada.  Study participants enrolled had to be > 18 years old and enrolled in ACLS.  Study participants attended an ultrasound training workshop to learn how to assess for carotid pulses via B-mode on a linear probe of an ultrasound.  Following the training and a three-hour washout period, participants were randomized either to start in the manual or ultrasound group as well as randomized to which live model (different than workshop models) and were made to identify the carotid pulse while being timed.  The entire study training period and experimental phase was completed on the same day.  The primary objective was to compare the speed of identifying carotid pulse checks with manual palpation versus ultrasound and the results revealed a mean pulse check time via ultrasound of 4.22 seconds and manual pulse check time of 4.71 seconds and no statistical difference between groups (pulse 90% CI: -1.77 to 0.39), however there was an overall higher variability/wider range of times in pulse detection with manual palpation which was statistically significant as well as a higher first-pass success rate with ultrasound of 110 attempts versus 95 with manual fist-pass attempts (p < 0.001 for both).

Strengths: The blinding was extremely well done.  For both manual and ultrasound recognition, each participant would begin their assessment after a countdown and then counted once again to the model’s heartbeat and this was all recorded and measured/timed by two independent blinded reviewers which I thought was brilliant.  

Limitations:  The researchers do not define what is positive on ultrasound exactly is; they loosely state in the discussion the lack of compressibility and pulsatile nature of the carotid artery but it does not seem as if these two findings were criteria that had to be specifically identified when checking for a pulse during the actual study.   

Areas of improvement:  Change the test population from models to patients in cardiac arrest versus a very controlled environment with volunteer models.  Or, one could get volunteers who represent the type of patient in cardiac arrest in terms of age and BMI. 

Conclusions:  

There is clearly utility for the role of POCUS for pulse checks in the setting of cardiac arrest, but both articles missed the mark somewhat.  

The first study had a lot lacking in terms of defining terms, methods, and calculations and the second study, while great in methodology, did no simulate cardiac arrest settings. 

Based on my critical appraisal of these articles, I will incorporate ultrasound into practice during cardiac arrest but be mindful of the limitations of the current research and need for higher quality trials in future research.

References:

Zengin S, et al.  Comparison of manual pulse palpation, cardiac ultrasonography and Doppler ultrasonography to check the pulse in cardiopulmonary arrest patients.  Resuscitation 2018;133:59-64.

Badra K, et al.  The POCUS pulse check: A randomized controlled crossover study comparing pulse detection by palpation versus by point-of-care ultrasound.  Resuscitation 2019;139:17-23.