Out-of-hospital Treatment of Opioid Overdoses in an Urban Setting


  • Karl A. Sporer MD,

    Corresponding author
    1. University of California, San Francisco; San Francisco General Hospital, Base Hospital: San Francisco, CA. Department of Emergency Services
    Search for more papers by this author
  • Jennifer Firestone BS,

    1. University of California, Los Angeles. School of Medicine, Los Angeles. CA
    Search for more papers by this author
  • S. Marshal Isaacs MD

    1. University of California, San Francisco; San Francisco General Hospital, Base Hospital: San Francisco, CA. Department of Emergency Services
    2. Department of Public Health. Paramedic Division, San Francisco‘, CA
    Search for more papers by this author

Emergency Services, Room IE21. San Francisco General Hospital, 1001 Potrero Avenue, San Francisco. CA 94110. Fax: 415–206–5818: e-mail: karlsporer@quickmail.ucsf.edu


Objectives: To investigate clinical outcomes in a cohort of opioid overdose patients treated in an out-of-hospital urban setting noted for a high prevalence of IV opioid use.

Methods: A retrospective review was performed of presumed opioid overdoses that were managed in 1993 by the emergency medical services (EMS) system in a single-tiered, urban advanced life support (ALS) EMS system. Specifically. all patients administered naloxone by the county paramedics were reviewed. Those patients with at least 3 of 5 objective criteria of an opioid overdose [respiratory rate <6/min, pinpoint pupils, evidence of IV drug use, Glasgow Coma Scale (GCS) score <12, or cyanosis] were included. A response to naloxone was defined as improvement to a GCS geqslant R: gt-or-equal, slanted14 and a respiratory rate geqslant R: gt-or-equal, slanted10/min within 5 minutes of naloxone administration. ED dispositions of opioid-overdose patients brought to the county hospital were reviewed. All medical examiner's cases deemed to be opioid-overdose-related deaths by postmortem toxicologic levels also were reviewed.

Results: There were 726 patients identified with presumed opioid overdoses. Most patients (609/726, 85.4%) had an initial pulse and blood pressure (BP). Most (94%) of this group responded to naloxone and all were transported. Of the remainder, 101 (14%) had obvious signs of death and 16 (2.2%) were in cardiopulmonary arrest without obvious signs of death. Of the patients in full arrest, 2 had return of spontaneous circulation but neither survived. Of the 609 patients who had initial BPs, 487 (80%) received naloxone IM (plus bag-valve-mask ventilation) and 122 (20%) received the drug IV. Responses to naloxone were similar; 94% IM vs 90% IV. Of 443 patients transported to the county hospital, 12 (2.7%) were admitted. The admitted patients had noncardiogenic pulmonary edema (n = 4). pneumonia (n = 2), other infections (n = 2), persistent respiratory depression (n = 2). and persistent alteration in mental status (n = 2). The patients with pulmonary edema were clinically obvious upon ED arrival. Hypotension was never noted and bradycardia was seen in only 2% of our presumed-opioid:overdose population.

Conclusions: The majority of the opioid-overdose patients who had initial BPs responded readily to naloxone, with few patients requiring admission. Noncardiogenic pulmonary edema was uncommon and when present, hypoxia was evident upon arrival to the ED. Naloxone administered IM in conjunction with bag-valve-mask ventilation was effective in this patient population. The opioid-overdose patients in cardiopulmonary arrest did not survive.