Friday, November 18, 2016

Title: Targeted Temperature Management After Pediatric Cardiac Arrest Due To Drowning: Outcomes and Complications

Authors: Moler FW, et al

Journal: Pediatric Critical Care Medicine, Aug 2016

This was a sub-group analysis of drowning patients from the Therapeutic Hypothermia After Pediatric Cardiac Arrest Out-of-Hospital (THAPCA-OH) Trial. This multi-center trial was a first of its kind randomized trial evaluating outcome in pediatric patients following cardiac arrest, comparing therapeutic hypothermia and normothermia. This study is in response to adult evidence over past decade showing improved neurologic outcome following cardiac arrest when therapeutic hypothermia is protocolized. Unfortunately, to date, the pediatric data has not shown this effect.

Study design: exploratory subgroup analysis

Study population: 

  • Pediatric patients 48 hours to 18 years of age
  • Cardiac arrest due to drowning
  • GCS < 5 on arrival (criteria from original trial)
  • 65 total patients analysed
    • 41 hypothermia group
    • 24 normothermia group
    • Median age 2.9 years, majority male
    • Bystander witnessed in 17%
    • Bystander CPR in 85%
    • Initial rhythm asystole in 65% and VF/VT in 3%
  • Targeted temperature management performed for 120 hours in all patients
    • Hypothermia group: 33C for 48 h, rewarmed over 16 hrs to 36.8C for remainder
    • Normothermia group: 36.8C maintained for duration
  • No difference in long-term neurologic outcome between groups
  • No difference in 12 month mortality between groups
  • These findings similar to outcomes from all-causes in the full trial
While these results unfortunately do not reveal any new hope for improving outcome in these patients, this trial at least adds evidence to an otherwise mysterious topic.  Currently, international guidelines call for initiating therapeutic hypothermia in drowning patients with cardiac arrest, but these recommendation are currently base on very little evidence.  


Friday, October 7, 2016

A population based study of drowning in Canada

Title: A population based study of drowning in Canada
Authors: Tessa Clemens , Hala Tamim, Michael Rotondi and Alison K. Macpherson
Journal: BMC Public Health

It is always great to see publications from friends of ours in the field of drowning research, whom we have come to admire over the years.  Tessa Clemens has done some great work through the Drowning Prevention Research Centre in Canada.  This study utilizes their robust database to study the epidemiology of drowning deaths in Canada from 2008 to 2012.

Study Summary

  • Retrospective database analysis, 2008-2012
  • Unintentional water-related fatalities
  • Primary findings
    • 2391 cases studied
    • Highest fatality rates in 65+ group
    • 82% of deaths were males
    • 75% of deaths in natural waters
  • Age groups
    • 0-4 year olds
      • high proportion of deaths related to falling in to pool or while bathing
      • More likely to be alone
    • 5-14 year olds
      • lowest risk for drowning death
      • high proportion of summertime and pool deaths
    • 15-19 year olds
      • Mostly while swimming or boating in summer
    • 20-35 year olds
      • 2nd highest fatality rate
      • 50% had consumed alcohol prior to event
    • 36-64 year olds
      • Tended to be boating
    • 65+ year olds
      • More likely to drown in bath tubs and pools in urban areas
The most interesting finding of the study was the high proportion of deaths attributed to the 65+ age group.  This is very different from data from other countries, including the US, where the 0-4 year age group has the highest fatality rate.  The study attributes the low rate, which is significantly reduced from previous Canadian studies, to successful public health campaigns. As with other studies that have come out of this group, what is most impressive is their ability to collect and track high-quality, country-wide data on drowning morbidity and mortality.

Monday, August 15, 2016

Clinical features and prognostic factors in drowning children: a regional experience

Title: Clinical features and prognostic factors in drowning children: a regional experience
Authors: Kyung Lae Son, MD, Su Kyeong Hwang, MD, Hee Joung Choi, MD, PhD
Journal: Korean Journal of Pediatrics, May 2016

This article applies the Szpilman Classification to retrospective drowning data in an attempt to determine prognostic factors.  The Szpilman Classificaiton was derived from research done by Dr David Szpilman in Brazil, and assigns a class (1-6) to drowning patients based on pre-hospital physical exam and vital signs.  In his original study, Dr Spzilman correlated these classes with outcome (the higher the class, the worse the outcome).

Study Summary

  • Study subjects
    • Patients experiencing submersion and admitted to study sites Jan 2005-Dec 2014
    • Total 29 patients (20 boys, 9 girls), all pediatrics
    • Szpilman grades
      • Grade 1-2: 6 patients
      • Grade 3-4: 3 patients
      • Grade 5: 13 patients
      • Grade 6: 7 patients
  • Outcomes
    • Poor neurologic (5 patients)
      • 2 in-hospital deaths (overall mortality 6.9%) 
    • Good neurologic (24 patients)
  • Prognostic factors
    • Age, sex, season, location had no prognostic value
    • Poor outcome
      • lower level of consciousness on arrival
      • higher Szpilman grade
      • All needed intubation and mechanical ventilation
      • Serum bicarbonate lower
      • No pupillary reflex
      • Hypotension
The findings of this study are not surprising. Clinical evidence of prolonged submersion and hypoxia was correlated with worse outcome.  This held true with the finding that a higher Szpilman grade was significantly associated with poor outcome.  The primary weakness of this study is the small study size, but, to my knowledge, it is one of the few if not the first to specifically use to Szpilman score to determine prognosis.


Thursday, July 28, 2016

Can we identify termination of resuscitation criteria in cardiac arrest due to drowning

Article: Can we identify termination of resuscitation criteria in cardiac arrest due to drowning: results from the French national out- of-hospital cardiac arrest registry

Authors: Hubert H, Escutnaire J, Michelet P, et al

Journal: Journal of Evaluation in Clinical Practice

The authors of this article set out to "identify prognostic factors for survival in order to determine advances cardiac life support (ACLS) termination criteria." They used  a French cardiac arrest registry which captures data from the country's EMS system, and evaluated multiple factors to produce a predictive model.

Design: Prospective, multi-center. Data from 7/1/2011 to 11/30/2014
Primary endpoint: patients' vital status at hospital admission


  • 32, 182 cardiac arrests recorded during study period
    • 291 cardiac arrests due to drowning (CAD)
    • 234 CADs analyzed for study
  • Patient and scene factors
    • median age 53
    • majority in rivers
    • median water temp 18 C (64 F)
    • 1/3 witnessed
    • 10% bystander BLS
  • Treatment factors
    • median "no-flow" time (cardiac arrest to first CPR): 10 minutes
    • only 1 patient received defibrillation
    • Majority of first recorded rhythms non-shockable
    • 6% had spontaneous circulation on EMS arrival
  • Prognostic Model
    • Five criteria with best prognostic power
      • Bystander ventilation
        • Those who receive it do better
      • EMS response time
        • Ther shorter, the better
      • Age
        • The younger, the better
      • No flow time
        • The shorter, the better
      • Location
        • Home (vs outside) did better
While I found the results of the study a little confusing, and the significance of the proposed criteria wasn't great, there are 2 things I like about this study:

1) A standardized, country-wide database was prospectively used to gather high quality data

2) Bystander BLS (primarily ventilation) far and away had the best prognostic effect. This further supports the fact that cardiac arrest from drowning is a problem of oxygen and its treatment must include reversal of hypoxia.


Wednesday, June 1, 2016

A call for the proper action on drowning resuscitation

Below is a copy of our joint Letter to The Editor to Resuscitation Journal which was recently published:

We ask for your attention to the following review of pathophysiology-based recommendations for the initial resuscitation of cardiac arrest caused by drowning, which will improve outcomes.
While compression-only CPR (CCPR) for the treatment of out of hospital sudden cardiac arrest (SCA) has gained great momentum, it has come at the cost of a misunderstanding as to the proper treatment for cardiac arrest due primarily to a non-cardiac cause. With SCA of presumed cardiac origin, the majority of which is the result of coronary artery occlusion, current out of hospital guidelines recommend CCPR due to the possibility of increasing bystander compliance.1 In cardiac arrest caused by a drowning event the cause of arrest is usually systemic hypoxeamia. In these cases CPR requires both ventilations and chest compressions aiming to reverse hypoxemia and provide coronary and cerebral perfusion.2
Because ventilations can be difficult to perform, more detailed CPR instruction following the traditional Airway–Breathing–Circulation (ABC) approach is required. This is especially important for: parents with pools; lifeguards; water sport athletes; and emergency healthcare providers. In addition peri-arrest drowning patients often have foamy material or emesis in their airway, which could easily dissuade a provider from providing ventilations if the importance of reversing hypoxemia is not understood. This is not a new concept, it has been included in resuscitation guidelines for many years, cited below, but sometimes is overshadowed by the push for CCPR that is used for SCA of a primarily cardiac origin.
European Resuscitation Council Guidelines for Resuscitation 2015. Section 2 and 4.3, 4
“Most cardiac arrests of non-cardiac origin have respiratory causes, such as drowning (among them many children) and asphyxia. Rescue breaths as well as chest compressions are critical for successful resuscitation of these victims.”3
“Most drowning victims will have sustained cardiac arrest secondary to hypoxia. In these patients, compression-only CPR is likely to be ineffective and should be avoided.”4
American Heart Association (AHA) 20105
“CPR for drowning victims should use the traditional A-B-C approach in view of the hypoxic nature of the arrest”5
“The first and most important treatment of the drowning victim is the immediate provision of ventilation.”5
For the reasons above, we ask all providers to educate themselves on the most recent recommendations in the 2015 ERC and 2010 AHA Guidelines for drowning resuscitation and their conclusions. The initial resuscitation of cardiac arrest caused by drowning should follow the ABC sequence; commencing with 2–5 initial rescue breaths followed by 30 chest compressions, and continue with 2 rescue breaths to 30 compressions.
These recommendations were jointly produced and are supported by the following organizations:
  • -
    Surfing Medicine International (Netherlands).
  • -
    Sociedade Brasileira de Salvamento Aquatico-SOBRASA (Brazil).
  • -
    Lifeguards Without Borders (USA).
  • -
    Surf Lifesaving Great Britain (UK).

  • References

  1. Kleinman, M.E., Brennan, E.E., Goldberger, Z.D. et al. Part 5: adult basic life support and cardiopulmonary resuscitation quality: 2015 American Heart Association guidelines update for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation. 2015; 132: S414–S435
  2. Szpilman, D., Bierens, J.J., Handley, A.J., and Orlowski, J.P. Drowning. N Engl J Med. 2012; 366: 2102–2110
  3. Perkins, G.D., Handley, A.J., Koster, R.W. et al. Adult basic life support and automated external defibrillation section Collaborators. European Resuscitation Council Guidelines for Resuscitation 2015: Section 2. Adult basic life support and automated external defibrillation. Resuscitation. 2015; 95: 81–99
  4. Truhlář, A., Deakin, C.D., Soar, J...., and Cardiac Arrest in Special Circumstances Section Collaborators. European Resuscitation Council Guidelines for Resuscitation 2015: Section 4. Cardiac arrest in special circumstances. Resuscitation. 2015; 95: 148–201
  5. Vanden Hoek, T.L., Morrison, L.J. et al. Part 12: cardiac arrest in special situations: 2010 American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation. 2010; 122: S829–S861

Tuesday, May 24, 2016

Predicting outcome of drowning at the scene: A systematic review and meta-analyses

Article: Predicting outcome of drowning at the scene:A systematic review and meta-analyses
Authors: Quan L, Bierens J, Lis R, Rowhani-Rahbar A, Morley P, Perkins G
Journal: Resuscitation, ePub ahead of print (May 2016)

One of the toughest questions to answer in the first few hours to days following the resuscitation of a drowning patient who remains comatose is what factors associated with the patient or the event can be used to determine prognosis.  While numerous studies have been done in the past to help answer this question, most are plagued by low quality data or designs and a focus on overall mortality instead of neurologic outcome. In this systematic review and meta-analysis, the authors focused on scene factors which could possibly help determine prognosis.


Study Type: Systematic review and meta-analysis
Period: 1979-2015
Inclusion: Cohort and case-control studies reporting:
  • Submersion duration 
  • Age
  • Water temperature
  • Salinity
  • EMS response time
  • Survival and/or neurological outcome

  • 24 cohort studies
  • Submersion duration strongest predictor
    • ≤ 10 min predicted high rate of good outcome
    •  ≥ 25 min associated with dismal outcome
  • Factors not showing prognostic value
    • Age
    • Water temperature
  • Factors showing some prognostic value
    • Salinity: salt water favorable
    • EMS response time: the shorter, the better
    • Both factors had very weak prognostic value

The most important conclusion of this paper, and one made many times before, is that the longer a person is submerged underwater, the worse chance they have for good neurologic outcome.  While there is no obvious cut-off time, and large variability, this analysis was able to produce a dose-response curve showing a correlation between submersion time and outcome. This time is a marker of anoxic brain injury.  The cut-offs of 10 min and 25 min were used to differentiate between strong chances of good outcome and bad outcome, respectively, but times in between are in a gray zone of uncertainty. An important note is that all of the papers analyzed were judge to have a low or very low level of evidence.

Monday, May 9, 2016

Wilderness Medical Society Practice Guidelines for the Prevention and Treatment of Drowning

We are very excited to announce the open-access publication of our recent practice guidelines developed for the Wilderness Medial Society. They cover the treatment of drowning, primarily in the austere and wilderness environment, including information to help guide the rural Emergency Medicine physician.

Click Here for link to article

Schmidt AC, Sempsrott JR, Hawkins SC, Arastu AS, Cushing TA, Auerbach PS.
Wilderness Medical Society Practice Guidelines for the Prevention and Treatment
of Drowning. Wilderness Environ Med. 2016 Apr 6.

Extracorporeal life support for victims of drowning

Title: Extracorporeal life support for victims of drowning
Authors: Burke CR, et al
Journal: Resuscitation, April 10, 2016 [epub ahead of publication]

This article analyses a subset of patients from the Extracorporeal Life Support (ELSO) international registry, which collects data from patients undergoing extracorporeal life support (ECLS) from over 400 international centers. Data for drowning patients from 1986 to 2015 were queried from the registry.

Primary outcome: survival to hospital discharge or transfer
Patients: 247 total (49 adults, 198 pediatrics)

  • 35% had cardiac arrest with ROSC prior to ECLS (Arrest group)
  • 31% ad ECLS initiated during cardiac arrest (ECPR group)
  • 34% did not have cardiac arrest before or during ECLS (No Arrest group)
  • Overall survival 51%
    • No Arrest: 74%
    • Arrest: 57%
    • ECPR: 23%
  • Younger patients more likely to be "No Arrest"
  • Non-survivors
    • more likely to be "Arrest" than survivors
    • No difference in age, sex, race
    • Presented with lower pH, lower SaO2, and more hypothermia

Unfortunately, this article does not answer any of the big questions in drowning resuscitation, primarily "who is walking out of the hospital with good neurologic outcome".  The authors acknowledge this weakness, but it is somewhat beyond their control as the registry itself is limited and primarily focused on ECLS itself and not specific diseases/injuries leading to ECLS.  I do like the use of an international registry to gather a wide-array of data across multiple institutions, but what is needed next is a method to match this data with more robust pre-hospital and hospital data and, most importantly, neurologic outcome data.  To data, the question as to whether ECLS is beneficial in drowning patients has yet to be answered, but registries like this may make it possible one day.  The conclusions gained from this article are in line with previous evidence: Those who experience cardiac arrest do worse, pediatrics tend to do better, and those who are severely acidotic, hypoxic, or hypothermic on arrival do worse.


Friday, April 1, 2016

Resuscitation of a Pediatric Drowning in Hypothermic Cardiac Arrest

Title: Resuscitation of a Pediatric Drowning in Hypothermic Cardiac Arrest
Authors: Dragann B, Melnychuk E, Wilson C, Lambert R, Maffei F
Journal: Air Medical Journal, March-April 2016

This article describes a fairly impressive case of prolonged submersion and severe hypothermia.  As with many similar cases recently described in the literature, the role of ECMO is described.  The unique thing about this case is that the patient regained a pulse while on the OR table to be cannulated for ECMO, and the procedure was never started.

Here are the case details

  • 22 month old missing in a snow storm in Pennsylvania
  • Found 30 minutes later submerged in 1.1 C water
  • Carried by neighbor to meet EMS crew
  • Ground EMS
    • Warm blankets, BVM, CPR
    • Initial rhythm PEA
    • IO access obtained, Epi given
  • Helicopter EMS (HEMS) dispatched to meet at hospital
    • CPR continued
    • Additional Epi and atropine given
    • Initial rectal temp 25 C (77 F)
    • HEMS team intubated on arrival
    • HEMS called nearby tertiary center with ECMO for transport
  • Arrived in Resus Bay 73 Minutes after initiation of CPR
    • Active rewarming
    • CPR continued
    • PEA rhythm
    • Initial pH 6.5
    • Additional Epi and bicarb given
    • End tidal CO2 12-20
  • Sent to OR for ECMO
    • Return of spontaenous circulation (ROSC) before cannulation
    • Total 101 minutes of CPR
    • Spontaneous respirations and some extremity movement
  • Transported to PICU
    • Continued active rewarming
    • Purposeful movements 8 hours after ROSC
    • Weaned from ventilation day 3
    • Discharged day 4
    • Day 9 follow up showed normal neuro status
This is quite an amazing case for many reasons: (1) very long CPR time before ROSC, (2) extremely acidotic (may be the lowest recorded pH in a survivor), (3) many many points in along the way where all efforts could have justifiably been aborted.  Currently, our best evidence (still not that great) says that there is little point in resuscitating patients who have been submerged > 30 minutes, although this to specific to water > 6 C.  In addition, the risk/benefit of flying a patient who has been in prolonged cardiac arrest greatly favors keeping the helicopter on the ground and ceasing efforts.  In addition, the patient's downtime, PEA rhythm, and severe acidosis all are very poor prognostic factors.  The few things working in the patient's favor were: very young and small, fall in very cold water with rapid cooling, surprisingly low initial potassium (studies show K > 10 poor prognosis), and fairly good end tidal CO2 (for a dead person) during resuscitation.

As with all cases of this nature, what happened with this patient is the exception, not the rule.  Most patients in this condition would likely have not survived or would have had nothing done to them given the poor chance of survival with good neurologic outcome.  And if some one survives it, they will likely have significant neurologic sequelae. A lot of things working for and against him, and obviously a very dedicated team and medical system. Very interesting case.