Intraosseous and Intravenous Epinephrine Administration Routes in Out‐of‐Hospital Cardiac Arrest: Survival and Neurologic Outcomes

Abstract

Background

The rate of survival after out‐of‐hospital cardiac arrest varies depending on the timeliness and effectiveness of prehospital interventions. This study was conducted to compare out‐of‐hospital cardiac arrest outcomes between intravenous and intraosseous routes and between upper and lower extremity routes for drug administration.

Methods and Results

We retrospectively analyzed data (collected using the Utstein template) from 1220 patients who had experienced out‐of‐hospital cardiac arrest in Taiwan's Taoyuan City between January 2021 and August 2023. The patients were stratified into intravenous and intraosseous groups by treatment approach and upper and lower extremity access groups by access site. The study outcomes were survival to discharge, favorable neurologic outcomes (Cerebral Performance Category score 1 or 2), and survival for >2 hours. The study groups were statistically compared before and after propensity score matching. Significant pre–propensity score matching differences were observed between intravenous and intraosseous groups, and the aforementioned study outcomes were better in the intravenous group than in the intraosseous group. However, the between‐group differences became nonsignificant after propensity score matching. Furthermore, lower extremity access and delayed epinephrine administration were associated with worse outcomes. Survival rates fell below 12.6% when time to treatment exceeded 15 minutes, particularly in the cases of intraosseous access and lower extremity access.

Conclusions

This study highlights the benefits of early intervention and upper extremity access for drug administration in patients with out‐of‐hospital cardiac arrest. Intraosseous access may serve as a viable alternative to intravenous access. Timely administration of essential drugs during resuscitation can improve clinical outcomes and thus has implications for emergency medical service training.

Nonstandard Abbreviations and Acronyms

AHA

American Heart Association

CPC

Cerebral Performance Category

HIO

humerus intraosseous

OHCA

out‐of‐hospital cardiac arrest

PSM

propensity score matching

TIO

tibia intraosseous

Clinical Perspective.

What Is New?

• We specifically investigated the differences in survival and neurologic outcomes when using upper versus lower extremity access for drug administration.

• The study also highlighted that delayed epinephrine administration and lower extremity access are linked to worse outcomes.

What Are the Clinical Implications?

• Early administration of epinephrine is crucial for improving survival rates in patients with out‐of‐hospital cardiac arrest, emphasizing the need for prompt prehospital intervention.

• When intravenous access is not achievable, intraosseous access remains a viable alternative, although it may be less effective if administered later or through lower extremity access.

• Emergency medical services training should emphasize mastering intravenous or intraosseous access techniques, especially in the upper extremities, and reducing delays in drug administration to enhance out‐of‐hospital cardiac arrest outcomes.

The incidence of out‐of‐hospital cardiac arrest (OHCA) currently ranges from 30.0 to 97.1 per 100 000 individuals, with a survival rate of 3.1% to 20.4%. ¹ , ² Prehospital resuscitation for OHCA relies on seamless implementation of the chain of survival, including dispatcher‐assisted cardiopulmonary resuscitation (CPR) and dispatcher‐assisted public access defibrillator use. In Taoyuan, Taiwan, the current (2024) rates of bystander CPR and public access defibrillator use are 71.1% and 8.7%, respectively. ³ With the improvement in the quality of basic life support, subsequent efforts must also be made to enhance advanced life support to strengthen the chain of survival.

Compared with placebo, epinephrine markedly improved the rate of survival to discharge of OHCA cases in the PARAMEDIC is the name of trial. ⁴ Hansen et al revealed that every minute of delay in administering epinephrine during OHCA reduces the likelihood of survival to discharge by 4% to 7%. ⁵ In addition, studies have indicated that early use of epinephrine in patients experiencing cardiac arrest leads to improved outcomes. ⁶ , ⁷ Current guidelines from the American Heart Association (AHA) recommend administering epinephrine as early as possible in patients with a nonshockable rhythm. ⁸

Establishing an intravenous route for administering drugs during the initial stage of resuscitation in patients with OHCA is a challenging task that requires a considerable amount of time ⁹ and can therefore delay drug administration. In 2021, emergency medical technicians (EMTs) in Taoyuan City, Taiwan, were authorized to use the intraosseous route for administering drugs in patients with OHCA. In a North Taiwan study, EMTs with 4 hours of relevant training proficiently used the intraosseous route for drug administration, enhancing its use during OHCA and facilitating timely drug administration in prehospital settings. ¹⁰

The 2020 AHA advanced cardiovascular life support guidelines indicate that the intravenous route is associated with better outcomes compared with intraosseous access, demonstrating better survival to discharge and neurologic outcomes in patients with OHCA. ⁸ The aforementioned studies have demonstrated that compared with intraosseous access, intravenous access has better outcome of survival to discharge and neurologic outcomes. ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ Therefore, the 2020 AHA advanced cardiovascular life support guidelines prioritize intravenous access and recommend intraosseous access only when intravenous access cannot be achieved. The evidence for this recommendation is classified as level 2a, but there is room for improvement, especially considering potential biases. The bias may be attributed to delay in administering epinephrine, difficulties in establishing intravenous access, or other characteristics associated with a higher risk of adverse outcomes. ¹¹

In recent studies, no significant differences between intravenous and intraosseous access on survival and neurologic outcomes were reported. ¹⁷ The tibial intraosseous (TIO) was associated with a reduced probability of achieving the return of spontaneous circulation and hospitalization in patients with OHCA. ¹² However, compared with humerus intraosseous (HIO), TIO is more reliable and accessible for operators managing critically ill patients. Therefore, TIO is also used as an alternative during resuscitation and first aid. ¹⁸

A consensus has yet to be reached on the optimal sites for establishing intravenous and intraosseous access. Common sites for achieving intraosseous access include the proximal tibia and proximal humeral head, whereas those for achieving intravenous access include the upper or lower extremities. Current evidence on the optimal sites for ensuring favorable outcomes in patients with OHCA is insufficient to enable definitive conclusions to be drawn. Thus, our study has 2 primary aims. The first is to compare survival outcomes of intraosseous and intravenous access. The second aim is to determine whether the outcomes in patients with OHCA differ between access established in the upper or lower extremities.

METHODS

The data and analytic methods of the study will be made available from the corresponding author upon reasonable request.

Study Design and Setting

This retrospective cohort study was conducted in Taoyuan City, Taiwan, between January 2021 and August 2023. The OHCA database from Taoyuan prospectively collected data based on the Utstein template. With a population of 2 309 770, Taoyuan City maintains an OHCA database encompassing data from its 13 first‐aid hospital emergency departments and 40 emergency medical services (EMS) units. According to Taiwan's Regulations of Emergency Medical Services, first‐aid hospitals provide 24/7 emergency care. This study identifies them as the nearest qualified hospitals capable of resuscitating patients with OHCA. Notably, patient data were anonymized before being included in the study.

In the EMS system of Taoyuan City, duty EMTs are classified as either intermediate‐level technicians or paramedics. Paramedics can perform intubation and administer epinephrine through the intravenous or intraosseous route. By contrast, intermediate‐level EMTs can only perform laryngeal mask airway insertion and establish intravenous or intraosseous access. Continuous mechanical CPR is provided to patients with OHCA after at least 2 minutes of manual CPR. In Taoyuan City, all EMS units follow the unified 2020 AHA advanced cardiovascular life support guidelines for resuscitation.

This study was approved by the Hospital Ethics Committee on Human Research of Taiwan's Chang Gung Medical Foundation (permit number: 202400158B0). After the study protocol was reviewed, the requirement to obtain informed consent was waived.

Selection of Participants

This study included adults who had experienced nontraumatic OHCA in Taoyuan City during the study period. We excluded patients having a do not resuscitate signed before cardiac arrest; those exhibiting clear signs of at‐scene death (decapitation, incineration, decomposition, or rigor mortis) or intoxication; those having OHCA because of drowning, hanging, or trauma; those who were pregnant; those with missing data (eg, information on prognosis or prehospital EMS record); and those not having a peripheral line established for epinephrine administration (Figure 1).

Figure 1. Flow diagram of the study period and selection of eligible subjects.

DNR indicates do not resuscitate; IO, intraosseous; and IV, intravenous.

The study conducted an investigation categorizing patients into intravenous and intraosseous groups either from upper or lower extremities by the first‐attempt access for medication administration. In the OHCA of Taoyuan City, the EMTs decide which peripheral route and medication to establish, so the EMTs in the scene determine the choice between intravenous or intraosseous. This study uses the first‐attempt access, which refers to the first route the EMTs decide to use in the scene, rather than the final successful route of medication administration, using an analysis as initially treated.

Data Collection

The collected information includes patient characteristics from both EMS run sheets and hospital medical records (eg, age, sex, event spot, witness status, bystander CPR, automated external defibrillator [AED] record, and medical history), number of EMT staff, transfer or nontransfer to a cardiac arrest center, site of peripheral line insertion, EMS parameters (response time and scene time interval), transport time, management time (time to epinephrine administration and time to AED use), and clinical outcomes.

The time to epinephrine administration (time to treatment) or AED use was defined as the interval between the time of the emergency call and that of epinephrine administration or AED use.

Outcome Measures

The primary outcomes were survival to discharge and favorable neurologic outcomes. The secondary outcome was survival for >2 hours. In Taiwan, the Fire Department considers survival for >2 hours as both sustained return of spontaneous circulation and a key quality indicator. Neurologic outcomes were evaluated using the Cerebral Performance Category (CPC) scale at discharge by physician attending. A CPC score of 1 indicates favorable cerebral performance (minimal or no neurologic impairment), whereas a CPC score of 2 indicates moderate cerebral disability (conscious and capable of independent function). Thus, CPC scores of 1 and 2 indicate favorable neurologic outcomes (good CPC).

Statistical Analysis

Categorical variables are presented in terms of numbers and percentages, and they were compared using a χ² test. Continuous variables are presented in terms of mean and SD values, and they were compared using the Student t test. Propensity score matching (PSM) was performed to adjust for the effects of potential confounding factors to equalize the probability of the intravenous or intraosseous access. The following confounders were included in the model in the calculation of propensity scores: age, sex, medical history, witnessed arrest, bystander CPR, event spot, EMT count, public AED use, initial AED rhythm, transfer or nontransfer to a cardiac arrest center, response time, scene time interval, transport time, time to epinephrine administration (time to treatment), and time to AED use. PSM (1:1) without replacement was performed using an 8‐to‐1 digit match algorithm, proceeding sequentially to the lowest digit match on the propensity score. Sample balance was assessed in terms of the standardized mean difference value, with an absolute value of >0.1 indicating a significant between‐group difference. After PSM, logistic regression was performed to evaluate the study outcomes. The analysis focused on the following outcomes: treatment groups (intravenous versus intraosseous), access site (lower versus upper extremities), and time to treatment initiation. Odds ratios (ORs) and adjusted ORs (aORs) with 95% CIs were calculated. The restricted cubic spline smooth function with 3 knots was used to visualize the association between the likelihood of the primary outcomes and time to epinephrine administration. Data analysis was primarily conducted using SAS, version 9.4. Additionally, Stata software, version 16.0 (StataCorp, College Station, TX), was used for implementing the restricted cubic spline function. Statistical significance was set at P<0.05 for the 2‐sided test.

RESULTS

Characteristics of Study Subjects

A total of 12 347 patients experienced OHCA in Taoyuan City during the study period. Among them, 6992 cases were not sent to the hospital because the patients were pronounced dead at the scene, and the family refused hospitalization with a signed do not resuscitate. Moreover, 3187 were excluded on the basis of our exclusion criteria (Figure 1). Finally, 2168 patients were included in this study. The patients' baseline characteristics are summarized in Table 1.

Table 1.

Baseline Characteristics of the Study Population Before and After PSM Analysis

Variables	Before matching			After matching
	Intraosseous (n=1323)	Intravenous (n=845)	SMD*	Intraosseous (n=840)	Intravenous (n=840)	SMD*
Age, mean±SD, y†	68.59±15.80	67.44±16.52	0.071	67.39±15.95	67.45±16.56	−0.004
Sex†			−0.073			−0.003
Male	871 (65.84)	585 (69.23)		581 (69.17)	582 (69.29)
Female	452 (34.16)	260 (30.77)		259 (30.83)	258 (30.71)
Medical history†
Diabetes	387 (29.25)	199 (23.55)	0.130*	180 (21.43)	199 (23.69)	−0.054
Hypertension	471 (35.60)	326 (38.58)	−0.062	321 (38.21)	321 (38.21)	0.000
Hemodialysis	349 (26.38)	209 (24.73)	0.038	230 (27.38)	209 (24.88)	0.057
Cancer	131 (9.90)	62 (7.34)	0.091	82 (9.76)	62 (7.38)	0.085
Witness†	469 (35.45)	377 (44.62)	−0.188*	359 (42.74)	372 (44.29)	−0.031
Bystander CPR†	904 (68.33)	617 (73.02)	−0.103*	620 (73.81)	612 (72.86)	0.022
Public AED use†	81 (6.12)	49 (5.80)	0.014	54 (6.43)	47 (5.60)	0.035
Event spot†			0.111*			0.054
Residential	1016 (76.80)	608 (71.95)		628 (74.76)	608 (72.38)
Public	307 (23.20)	237 (28.05)		212 (25.24)	232 (27.62)
Transport to†			0.034			0.057
CAC	91 (6.88)	51 (6.04)		63 (7.50)	51 (6.07)
Non‐CAC	1232 (93.12)	794 (93.96)		777 (92.50)	789 (93.93)
AED rhythm†			−0.041			0.006
Shockable	281 (21.24)	194 (22.96)		194 (23.10)	192 (22.86)
Nonshockable	1042 (78.76)	651 (77.04)		646 (76.90)	648 (77.14)
Site of peripheral line			−1.209*			−1.144*
Upper extremities	654 (49.43)	808 (95.62)		436 (51.90)	803 (95.60)
Lower extremities	669 (50.57)	37 (4.38)		404 (48.10)	37 (4.40)
No. of EMT† staff			0.028			0.008
≤3	39 (2.95)	21 (2.49)		22 (2.61)	21 (2.5)
>3	1284 (97.05)	824 (97.51)		818 (97.39)	819 (97.5)
Response time, mean±SD†	5.28±2.74	5.31±2.81	−0.009	5.30±2.55	5.31±2.81	−0.001
Scene time interval, mean±SD†	17.64±6.30	17.99±5.83	−0.059	17.73±6.41	17.98±5.83	−0.041
Transport time, mean±SD†	7.56±5.16	7.42±6.37	0.024	7.46±4.75	7.43±6.38	0.006
Time to treatment, mean±SD	19.93±7.62	21.80±7.61	−0.246*	20.04±7.60	21.80±7.63	−0.231*
Time to CPR, mean±SD†	9.83±10.86	9.68±4.31	0.018	9.72±9.23	9.68±4.31	0.005
Time to AED, mean±SD†	10.66±10.98	10.38±4.79	0.032	10.49±9.11	10.41±4.78	0.011

Values are expressed as number (percentage) and mean±SD unless otherwise specified. AED indicates automated external defibrillator; CAC, cardiac arrest center; CPR, cardiopulmonary resuscitation; EMT, emergency medical technician; PSM, propensity score matching; and SMD, standardized mean difference.

^*For assessing balance after matching, indicated SMD values (or absolute values >0.1) indicate imbalance.

^† Propensity score–adjusted variable.

The patients' mean±SD age was 68.14±16.09 years. Of the patients, 67.16% were men. The intraosseous group comprised 1323 patients (61%), whereas the intravenous group comprised 845 patients (39%). The intraosseous and intravenous groups differed significantly in terms of the proportion of patients with diabetes, rate of witnessed event, rate of bystander CPR, spot of OHCA occurrence, site of peripheral line, and time to epinephrine administration (time to treatment). The rate of witnessed event was higher in the intravenous group than in the intraosseous group (44.62% versus 35.45%). Similarly, the rate of bystander CPR was higher in the intravenous group than in the intraosseous group (73.02% versus 68.33%). On average, the time to treatment was 1.9 minutes longer in the intravenous group than in the intraosseous group.

After 1:1 PSM, 1680 patients were retained for analysis, with 840 in each treatment group. No statistically significant baseline characteristics between the intravenous and intraosseous groups were found with potential confounding factors. The access site for drug administration was the upper extremities, with 436 (51.9%) in the intraosseous group and 803 (95.65%) in the intravenous group. The time to epinephrine administration was 20.04±7.61 minutes for the intraosseous group and 21.8±7.63 minutes for the intravenous group.

Main Results

As presented in Figure 2, the likelihood of positive outcomes in all categories was higher in the intravenous group than in the intraosseous group: survival for >2 hours (254 [30.1%] versus 314 [23.7%]), survival to discharge (128 [15.5%] versus 144 [10.9%]), and favorable neurologic outcomes (92 [10.9%] versus 77 [5.8%]). After PSM, the between‐group differences were reduced. In multivariable logistic regression analysis (Table 2), the aORs for intravenous versus intraosseous outcomes showed increased probabilities for survival to discharge (aOR, 1.02 [95% CI, 0.72–1.35]), favorable neurologic outcome (aOR, 1.39 [95% CI, 0.94–2.05]), and survival >2 hours (aOR, 1.16 [95% CI, 0.90–1.49]), but these increases were not statistically significant. The aOR for access site in the lower extremities versus the upper extremities for outcomes, including survival to discharge (aOR, 0.32 [95% CI, 0.20–0.50]), favorable neurologic outcome (aOR, 0.27 [95% CI, 0.14–0.51]), and survival >2 hours (aOR, 0.65 [95% CI, 0.48–0.87]), were all statistically significant. The rate of survival to discharge decreased by ≈12% with every minute of delay in epinephrine administration (Table 2). We conducted a subgroup analysis on survival outcomes for upper and lower extremities, with results provided in Table S1. The probability of survival to discharge was visualized using the cubic spline approach. In Figure 3, for intraosseous and intravenous injections in the upper limb, survival decreased by 12.6% if treatment exceeded 15 and 16.8 minutes, respectively. Survival rates were consistently below average for lower limb injections. For intravenous injections in the upper limb, if the treatment time exceeded 16.3 minutes, favorable neurologic outcomes decreased by 7.8%. For the intraosseous group, this threshold was 15.7 minutes.

Figure 2. The distribution of survival to discharge, good CPC, and survival >2 hours before (A) and after (B) PSM.

A, The distribution of patients' survival >2 hours, survival to discharge, and good CPC, generated using R, is shown. Survival >2 hours and survival to discharge occurred in 251 (29.9%) and 127 (15.1%) of 840 patients in the IV group, and in 218 (26.0%) and 110 (13.1%) of 840 patients in the IO group after PSM. Good CPC (a score of 1 or 2) was observed in 91 of 840 patients (10.8%) in the IV group and in 58 of 840 patients (6.9%) in the IO group. The data are presented on a square root scale of the percentages of patients in each group by. B, The distribution of patients' scores on the CPC scale, ranging from 1 (minimal or no neurologic impairment) to 5 (death), was generated using R. In the IV group, scores of 1, 2, 3, 4, and 5 were observed in 40 (4.8%), 51 (6.1%), 21 (2.5%), 15 (1.8%), and 713 (84.8%) of 840 patients, respectively. In the IO group, these scores were observed in 39 (4.6%), 19 (2.2%), 23 (2.7%), 29 (3.5%), and 730 (87.0%) of 840 patients after PSM. The data are presented on a square root scale of the percentages of patients in each group. CPC indicates Cerebral Performance Category; IO, intraosseous; IV, intravenous; and PSM, propensity score matching.

Table 2.

Result of Multivariable Logistic Regression Analysis of Survival to Discharge, Good CPC, and Survival >2 Hours After PSM

Variables	Univariable		Multivariable
	OR (95% CI)	P value	aOR (95% CI)	P value
Survival to discharge
Treatment group (intravenous vs intraosseous)	1.18 (0.90–1.56)	0.234	1.02 (0.72–1.35)	0.918
Access site (lower vs upper extremities)	0.33 (0.22–0.50)	<0.001*	0.32 (0.20‐0.50)	<0.001*
Time to treatment	0.89 (0.87–0.90)	<0.001*	0.88 (0.86‐0.91)	<0.001*
Good CPC
Treatment group (intravenous vs intraosseous)	1.98 (1.44–2.71)	<0.001*	1.39 (0.94‐2.05)	0.098
Access site (lower vs upper extremities)	0.23 (0.12–0.41)	<0.001*	0.27 (0.14‐0.51)	<0.001*
Time to treatment	0.89 (0.86–0.91)	<0.001*	0.89 (0.86–0.91)	<0.001*
Survival >2 h
Treatment group (intravenous vs intraosseous)	1.22 (0.98–1.51)	0.073	1.16 (0.90–1.49)	0.254
Access site (lower vs upper extremities)	0.62 (0.48–0.81)	<0.001*	0.65 (0.48–0.87)	0.004*
Time to treatment	0.94 (0.93–0.96)	<0.001*	0.93 (0.92–0.95)	<0.001*

aOR indicates adjusted OR; CPC, Cerebral Performance Category; OR, odds ratio; and PSM, propensity score matching.

^*Statistical significance.

Figure 3. Probability of survival to discharge (A) and good CPC (B) versus time to treatment from a restricted cubic spline model.

A, A restricted cubic spline (red line) with 3 knots was used to visualize the association between the likelihood of primary outcomes and the time to epinephrine administration. The gray background represents the 95% CI. The blue dashed line represents the average probability of survival to discharge (12.6%). The intersection points with the restricted cubic spline (red line) are 15 minutes for the IO (upper limbs) group, 16.8 minutes for the IV (upper limbs) group, and 14 minutes for the overall data. B, A restricted cubic spline (red line) with 3 knots was used to visualize the association between the likelihood of primary outcomes and the time to epinephrine administration. The gray background represents the 95% CI. The blue dashed line represents the average probability of good CPC (7.8%). The intersection points with the restricted cubic spline (red line) are 15.7 minutes for the IO (upper limbs) group, 16.3 minutes for the IV (upper limbs) group, and 14 minutes for the overall data. CPC indicates Cerebral Performance Category; IO, intraosseous; and IV, intravenous.

DISCUSSION

In this study, the intravenous group showed better outcomes in good CPC and slightly higher likelihoods of survival to discharge and survival for >2 hours compared with the intraosseous group, but these differences were not statistically significant. Furthermore, outcomes for establishing access in the lower extremities were significantly poorer, with such access being associated with 68% and 73% reductions in the rates of survival to discharge and favorable neurologic outcomes, respectively. In addition, a time to epinephrine treatment of >15 minutes resulted in reduced survival rate and poorer neurologic outcomes in patients with OHCA.

During resuscitation with mechanical CPR and intubation, the treatment area in the upper body is more limited. Therefore, lower extremities are easier to implement intervention and may be reason why rate of intraosseous administration in the lower extremity is high. A local Taiwan study indicated that using the intraosseous route offers benefits such as convenience and rapid drug delivery. ¹⁰ However, we found a minimal time difference in drug administration between the intravenous and intraosseous groups, suggesting that the rapid access gained through the intraosseous route does not significantly improve prognosis in patients with OHCA.

Our cohort included patients for whom EMTs made onsite first‐attempt decisions about the use of the intravenous or intraosseous access. EMTs may predict potentially poor prognostic factors by their experience and think of that establishing intravenous access in these patients may be challenging.

Previous research on intravenous, HIO, and TIO suggests a better prognosis with upper extremity access, aligning with our findings. ¹⁹ , ²⁰ Other studies indicate that HIO access is a viable alternative to intravenous access. ²¹ , ²² , ²³ , ²⁴ However, in real‐world EMS scenarios, establishing intravenous access can be time‐consuming, potentially delaying critical interventions, such as epinephrine administration.

The 2020 AHA advanced cardiovascular life support guidelines referenced a study showing a positive trend toward intravenous access, differing from previous studies that found no significant difference in 30‐day survival. ⁸ , ¹¹ Establishing intravenous access can be challenging during resuscitation, often leading to the use of intraosseous routes. ²⁵ The advantages of intravenous access include lower vascular resistance and a higher rate of resuscitation drug distribution compared with intraosseous access. The drug concentration achieved through the TIO route is significantly lower. ²⁶ These findings suggest that intravenous access is preferable to intraosseous in patients with OHCA. ⁸ , ¹¹ In our study, intravenous access showed a slightly more favorable prognosis compared with intraosseous access in terms of survival and favorable neurologic outcome; however, the difference was not statistically significant. It is consistent with previous observational studies that demonstrated that drug administration through the intraosseous route typically results in poor outcomes of survival and neurologic outcome. ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰

Our findings revealed that compared with establishing TIO access, establishing HIO access significantly improved survival outcomes. This finding differs from previous studies that reported that TIO access is associated with a relatively high success rate and that there is no significant difference in survival outcomes between the HIO and TIO routes. ³¹ , ³² Therefore, our results suggest HIO should be performed before TIO during resuscitation.

We found that a delay in drug administration, regardless of the route of administration, is associated with lower rates of survival to discharge and favorable neurologic outcomes. Specifically, when the time to treatment exceeds 15 minutes, the rate of survival to discharge falls below 12.6%, with the survival rate decreasing by 12% with every minute of delay within the first 15‐minute period. Our results are consistent with previous studies that reported that early administration of adrenaline (within 10 minutes) in patients with OHCA can significantly improve the odds of survival to discharge, return of spontaneous circulation, and favorable neurologic outcomes. ⁶ , ³³ , ³⁴ These findings highlight the importance of early drug administration and emphasize the crucial role of prehospital intervention, given the time sensitivity of cardiac arrest events.

Our findings have major implications for EMS units. We discovered that drug administration through the upper extremities yields better outcomes than through the lower extremities. This finding can guide the development of effective modules for EMS training, particularly that for the establishment of the HIO route and an upper extremity peripheral line during manual or mechanical CPR in patients with OHCA. Our study underscores the need to prioritize the establishment of effective HIO or upper extremity intravenous access over that of lower extremity access during CPR. For favorable neurologic outcomes, after adjusting for access site and time to treatment, the aOR for the intravenous group compared with the intraosseous group was not statistically significant. This finding boost EMTs' confidence during treatment, as they can resort to HIO if intravenous access fails or is challenging to establish, ensuring continuous resuscitation efforts. Our study emphasizes the need for early establishment of a drug administration route in patients with OHCA. Therefore, Taiwan's prehospital EMS units must ensure that crucial interventions, such as epinephrine administration, are not delayed before hospital arrival, especially when the time from the emergency call to EMT arrival exceeds 15 minutes. Effective training on this topic for EMTs is essential.

Limitations

This study has some limitations. First, this was a single‐city study conducted in a city with a high population density. Thus, although the data may reflect the current state of the local EMS system, the generalizability of our findings is limited. Therefore, caution should be exercised when extrapolating the results to patients in other cities and countries. Second, although we performed PSM to adjust for confounders, such as patient demographics and medical history, we cannot eliminate the possibility of biases in the selection of peripheral routes by EMTs and variations because of their proficiency in establishing the intravenous or intraosseous route. Nevertheless, all involved EMTs were certified professionals who had received regular training. Finally, our aim in this study was to observationally compare treatment routes and access sites in terms of OHCA outcomes. Hence, we did not analyze postadministration drug concentrations; thus, our findings are not backed by pharmacokinetic data. Future studies are required to fill the indicated gaps in knowledge.

CONCLUSIONS

Our findings revealed no significant difference between the intravenous or intraosseous route for epinephrine administration in terms of primary outcomes in patients with OHCA. However, lower extremity access led to worse outcomes than did upper extremity access. HIO may serve as a viable alternative. Furthermore, a delay of >15 minutes in time to treatment significantly reduced the odds of survival to discharge and favorable neurologic outcomes.

Sources of Funding

This study was partly supported by Chang Gung Memorial Hospital (CMRPG1M0081 and CMRPG1N0081).

Disclosures

None.

Supporting information

Table S1