Background:

Foreign body airway obstruction (FBAO) is often described as an uncommon cause of Out of Hospital Cardiac Arrest (OHCA) accounting for approximately 1.4% of all OHCA. Reported incidents rates of FBAO causing cardiac arrest are unclear, and first aid by layperson are not well described.

The aim of the epidemiological part of the study is:

• to investigate information on actions taken by EMS-personnel and laypersons
• to investigate outcomes of hypoxic Cardiac Arrest due to foreign body airway obstruction in Denmark
• to increase overall survival. propose new guidelines and strategies to increase survival from OHCA caused by FBAO.

The aim of advanced text-string search algorithm part of the study is

- To investigate if an advanced text-string search algorithm can identify FBAO in medical records with high sensitivity

Methods:

National data will be collected from the verified 2016-2019 Danish OHCA register, and cases with FBAO prior to OHCA will selected via a direct marking by external validation and advanced text search. Patients reported as indisputably deceased (late signs of death) was excluded. Incidence rates per 100.000 citizens, survival rates to hospital and first aid actions by layperson are presented.

A pilot study have been conducted in regional data from 2016-2019 and the study group have concluded that, a national study is feasible with the current amount of data and the used methodology.

Expected outcome:

This study will enable targeted campaigns aimed at increasing survival from OHCA caused by FBAO. Potential campaigns might target the food items provided to potential vulnerable groups and guide focus for first aid recommendations. Further, with a deeper understanding of which airway management procedures most often are successful, it will be possible to improve EMS treatments of vulnerable groups.

Finally, a novel method of extracting information from the electronic medical records will be developed creating the foundation for future works on other prehospital conditions

Background Foreign body airway obstruction is a partial or complete blockage of airways, due to a foreign body (for example food, toys, coins etc.) Blocked airways will untreated result in cardiac arrest by hypoxia.

Previous research describes foreign body airway obstruction as an uncommon cause of Out-of-Hospital Cardiac Arrest (OHCA) accounting for approximately 1.1-1.4% of all OHCA. Reported incident rates of foreign body airway obstruction (FBAO) causing hypoxic cardiac arrest are unclear. It is reported to be as high as 11 per 100,000 person-years, comprising 10% of all OHCA, and neither Emergency Medical Services (EMS) treatment or first aid by layperson are well described. When organizing prevention campaigns for serious FBAO leading to cardiac arrest, knowledge on items with poor outcomes, vulnerable age-groups and treatments (appropriate and non-appropriate) initiated by both laypersons and EMS-personnel is essential. Previous studies suggest a bimodal distribution of FBAO compromised of the very young ingesting toys or other non-food related items (coins, balloons etc.) and elderly ingesting mainly food items . Several comorbidities are well documented to be over-represented in fatal cases in FBAO, notably cerebral infarction; dementia; schizophrenia; depression and Parkinson's disease .

However, FBAO leading to hypoxic cardiac arrest is a reversible cause that can easily be treated, and therefore prevented. Hence, hypoxic cardiac arrest due to FBAO provides a great potential for increasing overall survival from OHCA if efficient campaigns are implemented to provide the correct diagnosis, to facilitate prompt treatment from both EMS-personnel and laypersons. Such campaigns must be guided by evidence, realistic estimates of population and reliable baseline measures to evaluate effect.

Although, only minor studies present data from arrival to the hospital, several studies imply that the most common food items involved in FBAO is meat, bread, rice cakes, fruits and rice. There is a need to include prehospital data in a national study, to include relevant treatment administered by EMS or laypersons educated in first aid. A study by Igarashi et al. implies that the actions of laypersons has a significant impact on survival and that they, more often than not, use the Heimlich maneuver of abdominal thrust. However, only few patients in this study (12%) is treated by laypersons. More knowledge on the actions of bystanders are needed. The European Resuscitation Council currently recommend laypersons to deliver five back blows followed by an abdominal thrust/Heimlich maneuver if a person is suffering from severe FBAO (unable to cough). For health care personnel, including EMS, the MaGill forceps is first-choice when removing foreign bodies, with significantly higher survival rates reported. However, a complete report of all airway treatments and additional factors known to impact survival have not yet been reported.

There is approximately 5.400 OHCA annually in Denmark. The most updated 30-survival (2018) is 15,7%. This is among some of the highest in the world. A multitude of efforts and campaigns are targeting the different aspects of OHCA to improve survival. Although not yet verified, several sources report that approximately 3-4.5% of the entire Danish population (approximately 200.000-255.000 citizens) attend certified cardio pulmonary resuscitation (CPR) and automatic external defibrillator (AED) courses annually. These courses supposedly include recognition and treatment of FBOA.

Hence, with the relative few studies in the field, several knowledge gaps are persisting. There is a need to investigate who suffers from FBAO in order to create a profile to raise attention for health professionals responsible for treatment. Especially age and comorbidity has proven relevant in the few and limited studies in the field.

Most of these obstacles can be overcome with a national comprehensive investigation of prehospital data on OHCA caused by FBAO in a Danish setting.

Since the nationwide electronic medical record for the EMS providers was implemented in 2016, all OHCA have been registered electronically using the electronic medical records and been stored nationally. The electronic record contains a vast amount of information, including the medical professionals notes on each patient. This enables possibilities to conduct very direct advanced text searches for relevant trigger-words for identification of different subgroups of OHCA. This new approach opens for possibilities in targeting very specific subgroups, and extract information on diseases and treatment.

Aim

The aim of the epidemiological part of the study is:

• to investigate information on actions taken by EMS-personnel and laypersons
• to investigate outcomes of hypoxic Cardiac Arrest due to foreign body airway obstruction in Denmark
• to increase overall survival.
• propose new guidelines and strategies to increase survival from OHCA caused by FBAO.

The aim of advanced text-string search algorithm part of the study is

• To investigate if an advanced text-string search algorithm can identify FBAO in medical records with high sensitivity

Methods:

The study is a registry-based follow-up study of EMS patients in Denmark in 2016, 2017, 2018 and 2019 with cardiac arrest from foreign body airway obstruction as the presumed etiology.

OHCA data All cases of OHCA in which a resuscitative attempt is initiated is recorded in the Danish Cardiac Arrest Registry. Most variables have been collected since 2001 on paper sheets.

From 2016 and ongoing, the registry changed method of data-collection, moving from collecting paper-based OHCA records to extracting information on OHCA directly from the electronic medical records.

Because of the changeover, all records extracted from the electronic medical records was manually processed by a verification team reading the medical record and validating the information entered to the Danish Cardiac Arrest Registry. This ensured high-quality data from the approximately 5,400 annually registered OHCA in Denmark, within the period of 2016 to 2018. From the direct registration and later manual verification, several additional data sources are coupled with each registered OHCA. Notably, survival, initiation of bystander CPR and EMS actions.

Identifying FBAO This study investigates FBAO leading to OHCA. From the Danish Cardiac Arrest Registry all OHCA are collected and matched to the electronic prehospital medical record. This enables the possibility for advanced text searching in the entire prehospital medical record for cases with OHCA.

In a pilot-study performed at Copenhagen EMS in 2019, a method for identifying FBAO from trigger-words in the electronic medical record was completed. The pilot-study was conducted with data from the verified 2016-2018 Danish OHCA register on a population of patients treated by Copenhagen EMS. All records on patients with OHCA for a single year was reviewed, and OHCA from FBAO was manually identified. These records created the foundation for a list of initially 30 trigger-words specific for FBAO, that had been collected from the medical records in the note-fields, where ambulance physicians and paramedics could enter observations and descriptions of the patient and treatment as free text.

These trigger-words were structured and entered into a text-search algorithm able to search note-fields within the electronic medical records. When searching the electronic medical records from the OHCA register, the algorithm returned 121 records, consisting of OHCA presumably caused by FBAO. However, the initial search alone missed 17 of the manually registered FBAO leading to OHCA and conversely, FBAO not registered were identified from the search. After a manual review of the specified reports, new trigger-words were collected to include reports, not identified during the initial manual registration, ultimately leading to a revised search-string. This iterative process was performed, until no new reports appeared. After two search-string revisions, 273 OHCA presumably caused by FBAO were identified with no additional reports present. A total of 63 of these reports were verified manually.

From the pilot study, the investigators concluded that the method for identifying OHCA caused by FBAO, were feasible and efficient with a high specificity. With this method, the investigators can include national data and provide an essential base for future strategies aimed at increasing survival. However, the obstacle of manual verification across EMS regions persists and further data cleansing and revised search-strings are needed in order to provide high-quality data

In this study data will be collected from the verified 2016-2019 Danish Cardiac Arrest Registry. Cases with FBAO prior to OHCA will be identified with advanced text-searches as described above. With the increased amount of data, the investigators will be able to refine the algorithms, to achieve higher sensitivity and specificity. The remaining four regions not included yet, has all pledged data permission and agreed to initiate the study when funding have been acquired. The search-string will be modified and updated to include as many OHCA's with FBOA as etiology as possible. This process must be conducted in all of the four remaining regions separately.

Analysis data presentation Data will be pseudo-anonymized for analyses. The investigators will perform analyses only on an aggregated nationwide level. Descriptive statistics, fractions and annual incidence rates per 100.000 citizens, survival rates to hospital and first aid actions by layperson will be presented. General trends will be presented and Fisher's exact test together with odds ratios for survival will be calculated, stratified by age groups, bystander and EMS-actions. The final publication will be presented based on the STROBE statement for follow-up studies.

Data storage Data is stored on secure drive according to the regional instructions for safe conduct of data management.

Ethical considerations The investigators will follow GDPR and register the study at the Danish Data Protection Agency (Capital region of Denmark). As this is a Registry-based study, no ethical approval is required. However, the regional ethical committee will be contacted and presented with the current protocol to ensure proper ethical conduct.

Data access and analysis was approved by the Danish Data Protection Agency (reference 2007-58-0015, GEH-2014-019, I-suite 02737).

Perspectives This study will enable targeted campaigns aimed at increasing survival from OHCA caused by FBAO. Potential campaigns might target the food items provided to potential vulnerable groups and guide focus for first aid recommendations. Further, with a deeper understanding of which airway management procedures most often are successful, it will be possible to improve EMS treatments of vulnerable groups.

Finally, a novel method of extracting information from the electronic medical records will be developed creating the foundation for future works on other prehospital conditions

List of abbreviations CPR: Cardiopulmonary Resuscitation BLS: Basic Life Support EMS: Emergency Medical Services FBAO: Foreign Body Airway Obstruction OHCA: out-of-hospital cardiac arrest

Inclusion Criteria:

• All persons suffering from out-of-hospital cardiac arrest (OHCA) in Denmark from 2016-2020 identified in the verified danish OHCA register.

Exclusion Criteria:

• Patients reported as indisputably deceased (late signs of death) will be excluded.

Data will collected from the verified 2016-2019 Danish out-of-hospital cardiac arrest (OHCA) register, and cases with foreign body airway obstruction prior to OHCA will selected via a direct marking by external validation and advanced text search.