The need for multiple naloxone administrations for opioid overdose reversals: A review of the literature

Abstract

Background A growing challenge in the opioid epidemic is the rise of highly potent synthetic opioids, (i.e., illicitly manufactured fentanyl [IMF]) entering the US non-prescription opioid market. Successful reversal may require multiple doses of naloxone, the standard of care for opioid overdose. We conducted a narrative literature review to summarize the rates of multiple naloxone administrations (MNA) for opioid overdose reversal. Methods: A MEDLINE search was conducted for published articles using MESH search terms: opioid overdose, naloxone and multiple naloxone administration. Of the 2,101 studies identified, articles meeting inclusion/exclusion criteria were reviewed, categorized by primary and secondary outcomes of interest and summarized by data source and study design. Results: A total of 24 articles meeting eligibility criteria were included. Among EMS-based studies, MNA rates ranged from 9% to 53%; in general, bystander-reported studies were notably higher, from 16% to 89%. Variation in study design, data sources, year and geography, may have contributed to these ranges. Three studies that included longitudinal results reported a significant percent increase between 26% and 43% in annual MNA rates or a significant increase in mean naloxone doses over time (p < .001). Conclusions: This summary found that multiple naloxone administrations during opioid overdose encounters vary widely, have occurred in up to 89% of all opioid overdoses, and have significantly increased over time. Higher naloxone formulations may fulfill an unmet need in opioid overdose reversals, given the rising rates of overdoses involving IMF. Further studies are needed to gain a better understanding of MNA during opioid overdose encounters, particularly across a wider geographic region in the US in order to inform continuing efforts to combat the opioid epidemic.

Keywords:

• Naloxone
• Narcan
• multiple naloxone administrations
• dosing
• opioid overdose

Introduction

The opioid overdose epidemic remains a significant public health issue and continues to rise exponentially in the US. In 2017, there were 47,600 deaths due to opioid overdose, which represented 68% of all drug overdose deaths in the US.¹ In 2018, over 10 million individuals aged 12 or older (3.7%) in the US reported opioid misuse.² Recently, a new challenge has emerged—the introduction of synthetic opioids, including fentanyl, analogues such as carfentanil and precursor chemicals into the market. These synthetic opioids are highly powerful, with fentanyl and carfentanil estimated to be up to 100 and 10,000 times more potent than morphine, respectively.³ While fentanyl has been available in prescription form as an analgesic for some time, data indicates that the dramatic rise in drug overdose deaths in recent years can be attributed primarily to the influx of illicitly manufactured synthetic opioids.^4,5 The rate of US overdose deaths due to synthetic opioids increased from 9.0 per 100,000 population in 2017 to 9.9 in 2018, with 2 out of 3 opioid-related deaths involving synthetic opioids.^6,7 Data from 2013 to 2017 show the average annual percent increase was 69.8% in drug overdose deaths in the US due to synthetic opioids. This represents a marked increase compared to the average annual percent increase of 8.4% from 1999 to 2013.⁸ The most recent published estimates from January to June 2019 revealed that 61.5% of US drug overdose deaths were due to illicitly manufactured fentanyl (IMF), suggesting that the problem remains unabated.⁸

Additionally, the COVID-19 pandemic has further exacerbated the opioid overdose epidemic; recent data suggests an increase in non-prescription opioid use during the pandemic with the percent of tested individuals screening positive for IMF and heroin increasing by 35% and 44%, respectively.⁹

Approved by the FDA in 1971, naloxone is considered the standard of care and has been shown to be effective in opioid overdose reversals.^10,11 A key prevention strategy endorsed by the Centers for Disease Control and Prevention (CDC) and the American Medical Association (AMA) is to increase the distribution of and access to naloxone among bystanders, allowing bystanders to administer naloxone to reverse the overdose.^12,13 Since 1996, community-based harm reduction programs in the US have distributed naloxone directly to people who use opioids, and other bystanders.¹⁴ A recent analysis of data from the first half of 2019 found that nearly 40% of opioid and stimulant overdose deaths occurred in the presence of a bystander,¹⁵ thus take-home naloxone can be a critical, life-saving tool.

With the increase of synthetic opioids and the rapid onset of effect, evidence is emerging suggesting the need for increased doses of naloxone to reverse opioid toxicity. Numerous studies have been published examining the occurrence of multiple naloxone administrations (MNA) in opioid overdose reversals,^16–30 however, these studies have varied in study design, scientific rigor, sample size and geographic locations, making it difficult to obtain a comprehensive understanding of the frequency and trends of MNA occurrence in the US.

To our knowledge, a summary of the literature in this area has not been published. Thus, the purpose of this paper was to review the existing literature assessing the rates of multiple naloxone administrations for opioid overdose reversal to better understand the circumstances and context in which MNA have occurred. We sought to identify whether a stronger dose formulation of naloxone is needed and possible opportunities to improve the initial reversal response/outcomes for individuals experiencing an opioid overdose.

Methods

A narrative literature review was conducted to summarize published studies examining the rates of multiple administrations of naloxone for the reversal of opioid overdoses. A narrative review methodology was applied due to the dearth of published studies available in this area. Accordingly, results of this review were summarized qualitatively rather than applying meta-analytic summarization techniques.

Search strategy and article selection

A search was conducted in MEDLINE for peer-reviewed, original research articles published using the following MESH search terms: opioid overdose AND naloxone OR multiple naloxone administration. This initial search yielded 2,096 articles. Five additional papers were identified from the reference lists of the initially identified articles for a total of 2,101 articles. One reviewer reviewed all records and excluded 1,956 articles for the following reasons: studies not relevant to the topic (e.g., pharmacokinetic studies, clinical trials, treatments other than naloxone administered, opioid overdose detection tools, etc.) (n = 1,284), studies published before 2010 (n = 415), non-human studies (n = 385), and non-English language articles (n = 16). Full text articles were then assessed for eligibility. Studies that did not report either the number of naloxone doses administered or the total naloxone dose administered were excluded (n = 121). Due to the small number of available papers, case series and case study articles were included in this review. See Figure 1 for the flow chart depicting the article selection process.

Figure 1. Article selection schema.

Article abstraction

In accordance with the STROBE (Strengthening Reporting of Observational Studies in Epidemiology) guidelines, the following information from each article was entered into an evidence table to facilitate summarization: author/year, study design, analysis timeframe, sample size, study population, occurrence of multiple naloxone administrations (primary outcome of interest) and total dose administered (secondary outcome of interest) (see Table 1). Additionally, details regarding the rationale, objectives, key results, study limitations and generalizability of each article was documented and analyzed as part of the review process.

Table 1. Summary table of published studies on MNA.

Author, year	Study design	Analysis timeframe	Sample size	Study population	Proportion of cases with ≥2 doses of naloxone administered	Total dose administered
Faul, 2017^16	Retrospective database analysis of the National Emergency Medical Service Information Systems (NEMSIS) database	2012-2015	Number of patients who received naloxone by year 2012 N = 95,012 2013 N = 112,844 2014 N = 127,956 2015 N = 173,016	Individuals who received naloxone via EMS	Percentage of individuals receiving ≥2 doses of naloxone from EMS personnel: 2012:   13,765/95,012 (14.49%, 95% CI: 14.26–14.71%) 2013:   16,281/112,844 (14.91%, 95% CI: 14.71–15.11%) 2014:   20,884/127,956 (16.32%, 95% CI: 16.12–16.52%) 2015:   31,554/173,016 (18.24%, 95% CI: 18.06–18.42%) Overall, MNA increased 25.8% from 2012 to 2015	NR
Geiger, 2019^17	Retrospective database analysis of the NEMSIS database	2013–2016	Number of patients who received naloxone by year 2013 N = 107,666 2014 N = 122,506 2015 N = 166,002 2016 N = 206,354	Individuals who received naloxone via EMS	Significant increases (all p < .001) in percentage of individuals receiving ≥2 doses of naloxone from EMS personnel: 2013: 15% 2014: 16.5% 2015: 18.1% 2016: 21.4%	NR
Klebacher, 2016^18	Retrospective chart review of patients with suspected opioid overdose, treated with intranasal naloxone, based on EMR charts of largest provider of EMS services in New Jersey	April 2014–June 2016	N = 2,166	Individuals >17 years of age with suspected opioid overdose treated with intranasal dose of naloxone and subsequently managed by EMTs	195/2,166 (9%) received ≥2 naloxone doses 1971/2,216 (91%) experienced complete resolution with 1 naloxone dose • 53/2,166 (2.4%) of patients received 3 naloxone doses	NR
Fidacaro, 2019^19	A retrospective observational study of individuals receiving prehospital naloxone and transported to an inner-city ED in Camden, NJ	January 2, 2016–December 31, 2016	Total N = 473	Adults (≥18 years of age) treated with naloxone in a prehospital setting and transported by an inner-city hospital-based EMS to it’s affiliated ED	178/473 (38%) received multiple doses of naloxone in the prehospital setting • 34/473 (7%) received a repeat dose in the Emergency Department setting	NR
Purssell, 2020^20	A retrospective matched cohort study of opioid overdose cases treated in 2 urban Emergency Departments (ED) Patients who were administered low dose naloxone (LDN, ≤0.15 mg) were matched on a 1:4 ratio to patients who were administered high dose naloxone (HDN, >0.15 mg), based on initial respiratory rate	January 1, 2013 to December 31, 2017	Total N = 379 HDN n = 299 LDN n = 80	Patients with a presenting compliant of opioid overdose and were administered naloxone in the ED or administered naloxone by a layperson, paramedic or emergency personnel	199/379 (53%) received ≥2 doses of naloxone overall 1 naloxone dose used: 180 (48%) 2 doses used: 91 (24%) 3 doses used: 59 (16%) 4 doses used: 22 (6%) 5+ doses used: 27 (7%)	Mean (SD) naloxone dose for all subjects:   0.85 (0.76) Mean (SD) naloxone dose for all subjects until reversal: 0.58 (0.46) Mean (SD) naloxone dose for all subjects post reversal: 0.26 (0.44)
Maloney, 2020^21	A retrospective observational chart review of patients receiving prehospital naloxone and transported to Stony Brook University Hospital (SBUH), a suburban academic tertiary care center	January 1, 2014 to December 31, 2017	Total N = 513	Adults (≥18 years of age) treated with naloxone in a prehospital setting and transported by an ambulance to SBUH ED	160/513 (31%) received ≥2 doses of naloxone in the prehospital setting 1 naloxone dose used: 353/513 (69%) 2 doses used: 140/513 (27%) 3 doses used: 20/513 (4%)   41/76 (54%) received multiple doses of naloxone in the ED 19/76 (25%) were started on a naloxone infusion.	Median (range) of prehospital naloxone: 3 mg (2–3)
Krotulski, 2021^22	A pilot study of patients presenting to an urban tertiary care ED after a non-fatal opioid overdose.	June 2017 to December 2018	N = 20	Adult patients presenting to the ED who arrived after naloxone administration and who required venipuncture or intravenous access as part of clinical care.	10/20 (50%) received ≥2 doses of naloxone 1 naloxone dose used: 10/20 (50%) 2 doses used: 7/20 (35%) 3 doses used: 3/20 (15%)	Median (IQR) of calculated naloxone dose: 4 mg (2–5.1)
Bell, 2019^23	Prospective cross-sectional interviews of participants in the Prevention Point Pittsburgh (PPP) program. PPP is a community-based syringe access program that distributes naloxone to people who use drugs	January 2013–December 2016	N = 1,072	Participants in the PPP program who had used naloxone	462/1,072 (43%) had ≥2 naloxone doses administered 383/1,072 (35.7%) had 2 naloxone doses administered 47/1,072 (4.4%) had 3 naloxone doses administered 32/1,072 (3%) had 4 or more naloxone doses administered *no significant differences between any 4 years (2013–2016) on the average number of doses per overdose mean (SD) dose per reversal: 1.56 (0.93)	NR
Avetian, 2018^24	Telephone interviews with contacts of organizations that received 4 mg naloxone nasal spray to retrospectively collect data on circumstances regarding use of naloxone fo opioid overdose reversals	April–August 2016	Organizations contributing data N = 40 Opioid overdose reversals N = 261 Cases reporting doses administered N = 254	Participating organizations included law enforcement, county government, health departments, EMS, advocacy, community outreach and harm reduction organizations, volunteers, shelters and pharmacy service centers, located in CA, IN, ME, MD, NM, NC, & PA	89/254 (35%) had ≥2 naloxone doses administered 83/254 (33%) had 2 naloxone doses administered 5/254 (2%) had 3 naloxone doses administered • 1/254 (0.4%) had 4 naloxone doses administered	NR
Katzman, 2019^25	Prospective, 6-month cohort study to identify the characteristics of individuals participating in an opioid treatment program for OUD who used take-home naloxone to perform ≥1 opioid overdose reversal	April 2016–October 2016	Total N = 215 Used naloxone n = 44 Did not use naloxone n = 207	Adult participants being treated at University of New Mexico Addiction & Substance Abuse Program (ASAP)	37/65 (57%) received ≥2 doses of naloxone 1 naloxone dose used: 28 (43%) 2 doses used: 35 (54%) • 3 doses used: 2 (3%)	NR
Somerville, 2017^26	1) in-person interviews with individuals with opioid use disorder and 2) Retrospective medical examiner chart abstraction	Interviews: April 2016 Chart abstraction: Deaths occurred between October 2014–March 2015	Interviews: N = 64 Chart abstraction: N = 196	Interviews: Individuals ≥18 years of age who used illicit opioids in past 12 months who witnessed or experienced an opioid overdose in past 5 months Chart abstraction: Opioid overdose deaths occurring in 3 Massachusetts counties	• 40/48 respondents (83%) reported that ≥2 naloxone doses per suspected fentanyl overdose were used before person responded.	NR
Mahonski, 2019^27	A retrospective study of naloxone administrations from prepacked naloxone kits (PNK) reported to the Maryland Poison Control Center	January 1, 2015 to October 15, 2017	N = 1139 PNK administrations 2015 n = 121 2016 n = 433 2017 n = 585	Suspected opioid overdoses cases where either a layperson or law enforcement officer administered the naloxone from a prepacked naloxone kit	Percentage of individuals receiving ≥2 doses of naloxone: 2015: 19/121 (16%) 2016: 82/433 (19%) 2017: 118/585 (20%) Overall, MNA increased 25% from 2015 to 2017	Mean (median) dose increased significantly from 2015 to 2017 (p < .0001): 2015: 2.12 mg (2 mg) 2016: 2.54 mg (NR) 2017: 3.63 (4 mg)
Parkin, 2021^28	A qualitative interview study subsumed within a larger randomized trial evaluating 2 different overdose prevention training approaches in the NYC metropolitan area.	2014–2019	N = 56 Overdose events where > 1 dose of naloxone was administered for a opioid overdose event: N = 24	Individuals enrolled in the larger randomized trial that had witnessed an opioid overdose and had direct or indirect participation in an opioid overdose reversal using a take home naloxone kit in the past 3 months.	24/56 (43%) of overdose events where ≥2 doses of naloxone were administered Of overdose events where > 1 naloxone dose was administered, 10 (42%) involved administration of 2 doses of naloxone in under 2 min, outside of the recommended OEND response interval of 2–4 min.	NR
Schneider, 2021^29	A mixed methods, cross-sectional study design of individuals who used non-prescription opioids in Anne Arundel County, MD	November 2019–March 2020	Total N = 171 Received naloxone in past 6 months n = 61 Did not receive naloxone in past 6 months n = 110	Adults (≥18 years of age) who used non-prescription opioids in the past 6 months	54/61 (89%) received ≥2 doses of naloxone 1 naloxone dose used: 7 (12%) 2 doses used: 44 (72%) 3 doses used: 7 (12%) 4+ doses used: 3 (5%) Of those who had administered naloxone in the past 6 months, 23/29 (79%) reported that more than 2 doses were needed to reverse the overdose.	NR
Carpenter, 2020^31	Retrospective chart review of patients from a single Emergency Department (ED) and affiliated emergency medical services agency	January 2017–June 2018	Total N = 121 Opiates n = 28 Fentanyl n = 23 Opiates + fentanyl n = 70	Individuals with a diagnosis of opioid overdose, abuse or toxicity and had a urine drug screen (UDS) performed during hospitalization that was positive for opiates, fentanyl or both.	NR	Total naloxone dose for all subjects (median, range): 0.80 (0.18–5.20) Total naloxone dose for fentanyl (median, range): 0.80 (0.18–5.20) Total naloxone dose for opiates only (median, range): 0.58 (0.18–2.00) Total naloxone dose for opiates + fentanyl (median, range): 0.80 (0.18–3.60) *No significant differences in total naloxone dose administered between groups
Marco, 2018^32	Prospective survey of patients in the ED after an unintentional opioid overdose	July 2016–July 2017	N = 89	All adult individuals (≥18 years) presenting in the Miami Valley Hospital ED with an unintentional opioid overdose	NR	44/89 (49.4%) were administered 6 mg total naloxone dose or higher • The median prehospital dose was 4 mg (range: 0–26 mg)
Massey, 2017^30	Retrospective case series of an opioid outbreak in Cabell County, West Virginia	August 2016	N = 20 Probable cases n = 12 Confirmed cases n = 8	Individuals identified as probable opioid overdose cases within a 53-hour period in Cabell County, WV	• 6/20 (30%) received ≥2 naloxone doses	5 of 6 were IV administrations of 0.4 mg per dose
Sutter, 2017^33	Case series of 18 patients presenting with exaggerated opioid toxicity at a single hospital over an 8-day period	March 2016	N = 18	Individuals presenting to a single urban academic ED with symptoms of severe opioid overdose	NR	17/18 patients required boluses of naloxone and 4 required prolonged naloxone infusions (26–39 h)
Solomon, 2017^34	A case study of 1 female presenting to ED with severe opioid toxicity	N/A	N = 1	Adult female presented at large urban Ohio ED	NR	17 mg total naloxone (divided doses) were administered before patient returned to consciousness

NR: Not reported.

Results

A total of 24 articles met eligibility criteria and were included in this review. The articles were categorized by primary and secondary outcomes of interest and presented in order of data source and study design quality.

Proportion receiving ≥2 doses of naloxone—EMS

Figure 2 depicts the proportion of study participants who received ≥2 doses of naloxone by year of data collection for 13 of the articles that reported this metric with sample sizes greater than 20. Of these 13 studies, 6 were based on emergency medical services (EMS) or EMS and Emergency Department (ED) data, while 7 were based on bystander reported data. Two of the largest studies were retrospective analyses of the National Emergency Medical Service Information Systems (NEMSIS) database,^16,17 which contains pre-hospital data collected from EMS encounters across the US. Faul (2017) examined the proportion of opioid overdose events where ≥2 doses of naloxone were administered out of the total number of opioid overdose events where naloxone was used in the pre-hospital setting from 2012 to 2015. The findings showed a significant increase in the proportion of overdose events where ≥2 doses of naloxone were administered over time, with 14.5%, 14.9%, 16.3%, 18.2% from years 2012, 2013, 2014 and 2015, respectively. In this sample, nearly 1 out of every 5 patients received MNA to reverse the overdose in 2015.

Figure 2. Proportion of individuals receiving ≥2 naloxone doses for opioid overdose reversal by study.

A similar study by Geiger¹⁷ utilizing the same database measured the same outcome of ≥2 doses of naloxone administered out of the total number of opioid overdose events in the pre-hospital setting from years 2013 to 2016.¹⁷ Since the analysis timeframe of this study overlapped with Faul,¹⁶ nearly identical percentages of the proportion of opioid overdose events where ≥2 naloxone doses were administered were found from years 2013 to 2015. In 2016, the proportion was 21.4%, representing a significant increase over 18.1% from 2015. Both the Faul¹⁶ and Geiger¹⁷ studies were strengthened by having very large sample sizes representing EMS services in all regions within the US.

The above 2 NEMSIS studies were the only ones in this review that encompassed the entire US. However, the data did not include the administration route or the specific naloxone dose. Neither of these studies stratified the MNA rates by region, which would likely show significant variation, given the high incidence of synthetic opioid dispersion in certain regions of the country relative to others. By showing only the overall US rates, it is likely an underestimate of the true MNA occurrence in certain regions. Moreover, neither of these studies included prior administration of naloxone by a bystander in the calculation of the proportion receiving ≥2 doses of naloxone. While the proportion of individuals in both studies reported to have received naloxone from a bystander prior to EMS administration were small (<2%), this may be underreported, as EMS providers may not always be aware of, or record bystander interventions that occurred before their arrival.

The remaining studies that reported MNA for overdose reversals were conducted within limited geographic areas. An EMS-based study from a large EMS provider in New Jersey examined 2,166 adults who received naloxone for suspected opioid overdose from first responders between April 2014 to June 2016 and found the majority of cases (91%) were completely resolved after a single dose of intranasal naloxone. However, 142 (6.6%) did receive 2 doses of naloxone after lack of reversal from 1 dose, and 53 (2.4%) required a 3rd dose.¹⁸

Three additional retrospective studies that utilized both EMS and ED data from either urban or suburban tertiary care centers reported the proportion receiving MNA in both the prehospital and ED setting after a suspected opioid overdose. Fidacaro et al, 2019 reported that in January through December 2016, 178 of 473 (38%) patients received MNA in the prehospital setting and approximately 33 (7%) received an additional dose in the ED setting.¹⁹ In another study which utilized data from EMS and 2 urban ED care centers between January 2013 to December 2017, 199 of 379 (53%) patients received ≥2 doses of naloxone.²⁰ Further, 91 (16%) were administered 3 doses while 49 (13%) were administered ≥4 doses for reversal. Finally, the third study utilizing EMS and ED data from a suburban academic tertiary care center between 2014 to 2017 found 160 of 513 (31%) patients received ≥2 naloxone administrations in the prehospital setting, while 41 (54%) received multiple doses in the ED.²¹

It is noted that in each of these combined EMS/ED studies, sample sizes were relatively modest, participants were treated at either 1 or 2 tertiary care centers representing a relatively confined geographic area and generally did not include those who refused transport to the hospital upon ambulance arrival, thereby potentially limiting generalizability. It is also notable that MNA rates in studies utilizing EMS or combined EMS/ED data were generally low relative to bystander-based MNA rates in this review. EMS-based data may be reporting artificially lower MNA rates compared to other sources due to differing EMS guidelines for opioid overdose treatment by state or lack of reliable data on interventions performed prior to, or instead of EMS providers.

Proportion receiving ≥2 doses of naloxone—bystanders

Seven published studies reported the frequency of MNA based on self-reported data from bystanders who witnessed, administered or received naloxone for reversal of an opioid overdose event.^23–29 These studies collected data from 2013 up to 2020, thus provided more recent estimates of MNA and included laypersons who were often the first to administer naloxone.

One study that collected data as early as 2013 was based on self-reported MNA from individuals who participated in Prevention Point Pittsburgh (PPP), a community syringe access program that provides naloxone to people at risk for overdose.²³ Of the 1,072 opioid overdose reversals reported to PPP across years 2013 to 2016, 462 (43%) required ≥2 doses of naloxone.

Another study collected data from first responder and community-based organizations on naloxone administered in 261 overdose reversal events that occurred in from April to August 2016. Of the 254 overdose events that reported the number of naloxone doses administered, 35% required ≥2 doses of naloxone.²⁴

A third study that also collected data in the 2016 timeframe assessed MNA among participants enrolled in the University of New Mexico’s Addiction and Substance Abuse Program (UNM ASAP).²⁵ Of the 251 study participants who completed 3 and 6 month follow-up interviews from April to October 2016, 44 participants performed 65 opioid overdose reversals. Of the 65 overdose events, ≥2 naloxone doses were utilized in 37 (57%) events during the 6-month follow-up period.

A fourth study by Somerville (2017) interviewed 64 individuals in Massachusetts in April 2016 who had witnessed a suspected IMF overdose in the past 5 months. This study found that 48 (75%) witnessed a suspected IMF overdose, and reported that ≥2 naloxone doses were administered in 40 (83%) of these overdose events.²⁶

More recently, three additional studies reported the proportion of MNA based on bystander self-report up to the 2020 timeframe. Mahonski assessed the rate of MNA among laypersons (including law enforcement officers [LEOs]) who were provided naloxone kits through an opioid overdose education and naloxone distribution (OEND) program in Maryland from 2015 to 2017.²⁷ Over 3 years, the proportion receiving ≥2 doses of naloxone rose from 16% (19/121) in 2015 to 20% (118/585) in 2017, representing a 25% increase. This corresponded with a significant increase in the mean naloxone dose from 2.12 mg (median: 2 mg) to 3.63 mg (median: 4 mg) during the same time frame (p < .0001). Notably, this study relied on individuals to report the overdose event to a regional poison control center on their own, which the authors noted were predominantly based on LEO-reported events. Thus the relatively low MNA rates may have been due to the lack of voluntary layperson reports.

A qualitative interview study that assessed the impact of 2 OEND training programs among individuals who use opioids found that between 2014 and 2019, 24 of 56 (43%) opioid overdose events included in the sample involved ≥2 naloxone administrations.²⁸

The most recent bystander self-report study in this review included 173 adults who used non-prescription opioids in the past 6 months between November 2019 and March 2020. This study reported that 54 of 61 (89%) patients received ≥2 doses of naloxone.²⁹ Of the participants who administered naloxone to someone else, 23 of 29 (79%) reported that more than 2 doses were needed to reverse the overdose. Notably, the most recent data collected from bystanders showed the highest rates of MNA, which may reflect the rising rates of IMF in the drug supply. Considering that the typical take home naloxone kit contains 2 doses of naloxone, these rising rates of MNA suggest the kits may need more naloxone to ensure an amount sufficient to reverse the overdose.

It is important to note that the generalizability of these bystander studies may be limited as each study had relatively small sample sizes, were conducted with participants within a defined geographic region and relied on bystander recall and self-report. However, despite these limitations, these 7 bystander studies offers further evidence of the increasing frequency of MNA.

Total naloxone doses administered

Two published studies of naloxone use for opioid overdose events reported the total naloxone dose, rather than number of doses. The first study was a single-center, retrospective chart review study that included both EMS and hospital EMR data from January 2017 to June 2018.³¹ Total naloxone dose was compared between 3 study groups: IMF only (n = 23), opiates only (n = 28) and IMF + opiates (n = 70). Assessment of the total dose administered revealed that nearly 40% of the sample received a higher total naloxone dose over 0.4 mg. The median total dose did not differ significantly between the IMF only (0.80, 0.18–5.20) IMF + opiates (0.80, 0.18–3.60), and the opiates only (0.58, 0.18–2.00) involved group.³¹ Limitations included the use of a qualitative, rather than quantitative urine immunoassay as a surrogate of opiate and IMF exposure by which the study groups were compared and the conversion of naloxone doses delivered via non-IV routes to an IV dose equivalent using literature-based assumptions, which may have lacked accuracy.

A study by Marco (2018) consisted of a questionnaire interview, supplemented by EMR data of adults presenting in the ED with an opioid overdose between July 2016 through July 2017.³² Of the 89 participants, most (88%) received naloxone in the pre-hospital setting only. The mean prehospital naloxone dose administered was 5.8 mg (±5.0), and the mean total naloxone administered was 5.9 (±4.9).

Case series or studies

Four case series or case studies with sample sizes of ≤20 that reported MNA frequency were also included in this review. A retrospective public health investigation of an opioid overdose outbreak in West Virginia in August 2016 found that 6 of 20 (30%) events required ≥2 doses of naloxone. Of these, 5 were IV administrations of 0.4 mg per dose.³⁰ In a case series of 18 patients who presented with opioid toxicity at a single hospital in March 2016, 17 of 18 cases required ‘boluses of naloxone’ suggesting multiple doses were required for reversal.³³ A pilot study of 20 patients presenting to an urban tertiary care ED after a non-fatal opioid overdose found 50% received ≥2 doses of naloxone.²² A final case study reported on a female presenting to the ED with severe opioid toxicity who required 17 mg total naloxone dose for reversal indicating MNA.³⁴

Discussion

Multiple naloxone administrations

Overall, this literature review found that the reported rates of MNA varied widely and have occurred in up to 89% of opioid overdose encounters in the US, were exceptionally high in certain regions, and have increased significantly over time. Among studies that utilized EMS data, MNA rates ranged from 9% to 53%,^16–22 while studies based on bystander-reported data were notably higher, ranging from 16% to 89%.^23–29 The wide range in reported MNA rates overall is not surprising given the significant variation in study design and attributes, the different data sources utilized, the year and geography studied, and the difficulty and limitations in accurately capturing this data. Most of the studies had relatively small sample sizes and were conducted in a limited geographic region, with only 2 studies based on national data with robust sample sizes. Thus, it is difficult to extrapolate an accurate rate of MNA for opioid overdose reversals in the US. However, given that lower MNA rates were generally reported in the EMS-based studies relative to the bystander-reported studies, there may be subtleties in the data sources or other factors that may explain differing MNA rates. For instance, lower MNA rates found in EMS-based studies may reflect variation in EMS protocols by state or EMS agency. Some states provide specific guidance on the maximum naloxone dose that should be administered, while other states recommend titrating naloxone until respiratory depression is resolved, but not necessarily to restore full consciousness.^9,35,36 Additionally, bystander interventions may not always be reported to EMS providers upon their arrival nor recorded by EMS, thus bystander administration might not be reliably recorded in EMS systems. EMS systems will also not include opioid overdose events that are treated and resolved solely by bystanders. Due to concerns about harassment or arrest, bystanders may not report prior naloxone administration to EMS or remain present when EMS arrives. For these reasons, EMS-based data may be reporting artificially lower MNA rates compared to other data sources that do not have these issues.

Association of IMF with multiple naloxone administrations

This review found very limited documentation on how rapidly IMF use in the real-world results in overdose. One study reported that in 36% of cases out of the 125 opioid overdose deaths, there was evidence an overdose occurring within seconds to minutes after drug use, and 90% were without a pulse upon EMS arrival.²⁶

This evidence of rapid IMF overdose onset, albeit limited, suggests that having a bystander who can administer enough doses of naloxone to reverse the overdose may be a critical factor in overdose survival. Given the increasing rates of bystander administration combined with the rising rates of IMF availability and the rapid onset of effect, bystanders should have training and access to enough naloxone to administer during the critical time interval before EMS services arrive to increase the likelihood of survival.

A recently developed opioid receptor quantitative systems pharmacology model simulated the effects of a single dose of 2 mg, 5 mg and 10 mg intramuscular (IM) naloxone dosage formulations on 3 different levels of fentanyl exposure (25 ng/ml, 50 ng/ml, 75 ng/ml).³⁷ When simulating the time predicted to reduce µ-receptor occupancy by fentanyl to below 50%, a trend emerged whereby increasing the naloxone dose resulted in faster time to achieve this threshold. The model predicted it would take 10 min to reduce the fentanyl receptor occupancy by 50% with a 2 mg IM (or 4 mg intranasal) naloxone dose based on 50 ng/ml level of fentanyl exposure, which the authors noted may be too long to successfully reverse the overdose. However, with higher naloxone dose administrations, the time to reduce fentanyl receptor occupancy to 50% was less than 5 min, suggesting that due to increased fentanyl exposure, higher doses of naloxone may be warranted.

Opioid withdrawal syndrome (OWS)

A concern with increasing naloxone dosing is the likelihood of an individual experiencing OWS precipitated by an abrupt reversal of the overdose. Only 5 studies in this review documented OWS or other adverse events related to naloxone and reported a range from 0.6% to 38%.^{20,21,24,27,31} A recent systematic literature review evaluating naloxone dosing and adverse events (AEs) found that only half of the reviewed studies (86/174) reported AEs, and among these, 11% of patients experienced OWS.³⁸ The lack of documented cases of OWS following increased naloxone dosing reported could suggest that the risk is low, but it may also reflect the difficulty in capturing this metric, thus more research is needed to better understand the likelihood of OWS with increased naloxone dosing. However, given the life-threatening consequences of administering inadequate dosing for reversal, and the context of unknown IMF exposure, the risk of an unsuccessful reversal may outweigh the risk of OWS.

Other considerations

Finally, an unprecedented global health threat, the COVID-19 pandemic, has further exacerbated an already challenging opioid crisis. A recent study of a national sample of 872,762 urine specimens, which compared the period between the pre-COVID-19 baseline (January 1, 2019–March 14, 2020) to the COVID-19 pandemic period (March 15–May 16, 2020), reported a 35% (p < .01) significantly increased positivity rate for IMF during the COVID-19 period.³⁹ As the COVID-19 pandemic persists, it is expected that IMF will continue to rise which will act as a further catalyst to the pre-COVID-19 opioid overdose rate of progression.

Limitations

Several limitations to this review are noted. First, a systematic literature review (SLR) was not conducted due to the small number of published studies examining the frequency of MNA for opioid overdose reversals. All published studies that we could find that reported MNA for opioid overdose reversal were included in this review, regardless of study robustness or small sample sizes.

Another limitation was the paucity of large, national studies examining MNA. Most of the studies in this review had small samples sizes from relatively limited geographic areas, which may not be necessarily representative of the US. Two studies utilized the NEMSIS data, which is a large, national database of EMS encounters. It should be noted however that there are issues with representativeness even with the large NEMSIS database, as there is variation among states in the selection criteria used to submit EMS data to NEMSIS.⁴⁰ Furthermore, the two studies in this review that utilized this database did not include bystander administration of naloxone prior to EMS arrival in their calculation of MNA, thereby potentially underestimating the true rates of MNA.

Conclusion

This summary found that multiple naloxone administrations during opioid overdose encounters have increased over time and appeared to be exceptionally high in certain US regions. Lower rates of MNA reported in EMS-based studies relative to bystander-based studies may partially reflect variation in EMS guidelines by state and data limitations. Overall, evidence is emerging that higher formulations of naloxone may fulfill an unmet need, particularly given the increasing rates of IMF exposure. Further studies are needed to gain a better understanding of more recent occurrences of MNA during opioid overdose encounters, particularly across a wider geographic region in the US in order to inform continuing efforts to combat the opioid epidemic.

Authors’ contributions

Randa Abdelal, A. Raja Banerjee, Neyla Darwaza, Diane Ito and Josh Epstein were involved in the conception and design of this review, analysis, interpretation of the results and drafting/revising the paper for intellectual content. Suzanne Carlberg-Racich was involved in the interpretation of the results and drafting/revising the paper for intellectual content. All authors were involved in the final approval of this manuscript to be published and agree to be accountable for all aspects of the work herein.

Disclosure statement

Suzanne Carlberg-Racich has received consultancy fees from Hikma Community Health, a Subdivision of Hikma Specialty USA Inc. A. Raja Banerjee, Randa Abdelal, and Neyla Darwaza have received stock options exercisable for, and other stock awards of, ordinary shares of Hikma Pharmaceuticals in the course of their employment at Hikma Pharmaceuticals. Diane Ito and Josh Epstein are employees of Stratevi, a consulting firm that has received research funding from Hikma Community Health, a Subdivision of Hikma Specialty USA Inc., to conduct this review.

Additional information

Funding

This review was funded by Hikma Community Health, a Subdivision of Hikma Specialty USA Inc. The funding organization was involved in the design of the study; interpretation of the data; preparation, review, and approval of the manuscript; and decision to submit the manuscript for publication.