Secondary Logo

The Association of Hospital Rate of Delayed Epinephrine Administration With Survival to Discharge for Pediatric Nonshockable In-Hospital Cardiac Arrest

Raymond, Tia T., MD1; Praestgaard, Amy2; Berg, Robert A., MD3; Nadkarni, Vinay M., MD4; Parshuram, Chris S., MB ChB, PhD5 for the American Heart Association’s Get With The Guidelines-Resuscitation Investigators

Pediatric Critical Care Medicine: May 2019 - Volume 20 - Issue 5 - p 405–416
doi: 10.1097/PCC.0000000000001863
Feature Articles
Free
SDC

Objectives: To evaluate the variation of hospital rates of delayed epinephrine administration in pediatric patients with nonshockable in-hospital cardiac arrest, and the association of those rates with event, 24-hour, and overall survival to hospital discharge.

Design: A retrospective evaluation was performed. Delayed epinephrine was defined as greater than 5 minutes between the time the need for chest compressions was identified and epinephrine was administered. The main outcome was the association of hospital rate of delayed epinephrine administration with survival to hospital discharge. Secondary outcomes were event and 24-hour survival. Evaluation used hierarchical logistic regression and included 13 patient/event-level and seven hospital-level factors.

Setting: Hospitals with greater than 6 months data in the American Heart Association’s Get With the Guidelines—Resuscitation registry (2000–2016) and greater than or equal to five total pediatric cardiac arrests with nonshockable rhythm.

Patients: Children less than 18 years old with index nonshockable in-hospital cardiac arrest treated with greater than or equal to one epinephrine dose.

Interventions: None.

Measurements and Main Results: One-thousand four-hundred sixty-two patients at 69 hospitals were included: 218 patients (14.9%) had epinephrine delay rates ranging from 0% to 80% of events (median, 15.6%; interquartile range, 7–25%). The median and interquartile range of hospital level delay was 16% (7–25%). Patient/event-level predictors of delayed epinephrine were asystole (odds ratio, 1.54 [95% CI, 1.10–2.16]) and insertion of an endotracheal tube (odds ratio, 1.86 [95% CI, 1.27–2.73]). Hospital size less than 200 compared with greater than or equal to 500 beds (odds ratio, 3.07 [95% CI, 1.22–7.73]) and ICU location (odds ratio, 0.51 [95% CI, 0.36–0.74]) were associated with epinephrine delay rates. After adjustment, increasing quartiles of epinephrine delay were associated with lower patient and hospital-level return of spontaneous circulation (p = 0.019, p = 0.006) and 24-hour survival (p = 0.018, p = 0.002) respectively, but not survival to discharge (p = 0.20, p = 0.24).

Conclusions: Delayed epinephrine administration following pediatric nonshockable in-hospital cardiac arrest varies significantly between hospitals. Hospitals with higher rates of delayed epinephrine administration had worse patient- and hospital-level outcomes after adjusting for multiple patient- and hospital-level factors. Delayed epinephrine administration may directly contribute to increased mortality risk and/or may be a marker of unmeasured elements of hospital resuscitation performance.

1Division of Cardiac Critical Care, Department of Pediatrics, Medical City Children’s Hospital, Dallas, TX.

2Department of Biostatistics, University of Pennsylvania, Philadelphia, PA.

3Department of Anesthesiology and Critical Care Medicine, The Children’s Hospital of Philadelphia, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA.

4Department of Anesthesiology, Critical Care and Pediatrics, Center for Simulation, Advanced Education and Innovation, Center for Resuscitation Science, The Children’s Hospital of Philadelphia, Philadelphia, PA.

5Division of Critical Care Medicine, Department of Pediatrics, Hospital for Sick Children, Toronto, ON, Canada.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (http://journals.lww.com/pccmjournal).

Members of the American Heart Association's Get With The Guidelines Resuscitation Investigators can be viewed in the Acknowledgments.

Supported, in part, by the Congenital Heart Surgical Unit at Medical City Children’s Hospital.

Dr. Praestgaard’s institution received funding from Medical City Children’s Hospital. Dr. Parshuram disclosed that he holds shares in a clinical decision support company in part owned by the Hospital for Sick Children, but no money has been paid to him. The remaining authors have disclosed that they do not have any potential conflicts of interest.

For information regarding this article, E-mail: tiaraymond@me.com

Approximately 16,000 children in the United States have a cardiac arrest each year, predominantly in a hospital setting (1, 2). An initial rhythm of pulseless electrical activity or asystole (i.e., nonshockable rhythm) is most common and is associated with significant mortality, with only 25–40% of patients surviving to hospital discharge (1, 3–5). Despite efforts in resuscitation research and improvement in outcomes after in-hospital pediatric resuscitation during the last 30 years (4–6), there are few evidence-based interventions besides supportive care for pediatric cardiac arrest patients with a nonshockable rhythm (6, 7).

The most recent 2015 American Heart Association (AHA) Guidelines Update for Cardiopulmonary Resuscitation (CPR) and Emergency Cardiovascular Care by the “International Liaison Committee On Resuscitation Pediatric Life Support Task Force” stated that it is reasonable to consider administering a 0.01 mg/kg dose of IV/intraosseous epinephrine every 3–5 minutes during pediatric cardiac arrest (6). A recent report found that delay in epinephrine administration for adult in-hospital, nonshockable cardiac arrest was associated with decreased chance of return of spontaneous circulation (ROSC), survival to discharge, and good neurologic outcome (8). Similarly, a recent multicenter cohort study of pediatric in-hospital cardiac arrest (IHCA) reported that delayed epinephrine administration was associated with a decreased chance of ROSC, 24-hour survival, survival to hospital discharge, and survival to hospital discharge with a favorable neurologic outcome among patients with an initial nonshockable rhythm (9). It is conceivable that hospital efforts aimed at improving timely administration of epinephrine could potentially improve survival in this population where survival rates have been traditionally poor.

Among children with IHCA, the extent of hospital variation in delayed epinephrine administration and its impact on hospital survival rates is largely unknown, a critical step toward sharing best practices so that all centers can improve their care of patients with cardiac arrest. To address this gap in knowledge, we used data from the Get With The Guidelines–Resuscitation (GWTG-R) registry to examine hospital-level variation in rates of delayed epinephrine administration. We then examined the implications of hospital variation by determining the association between a hospital’s rate of delayed epinephrine administration and its rate of overall survival for pediatric patients with nonshockable IHCA. We hypothesized that there will be substantial variation among hospitals in rates of delayed epinephrine administration for pediatric IHCA due to a nonshockable rhythm and that hospitals with higher rates of delayed epinephrine administration will have lower rates of survival. An improved understanding of current hospital performance in epinephrine administration times and factors associated with delayed epinephrine administration is critical for developing effective interventions that could be implemented at hospitals to improve outcomes after pediatric IHCAs.

Back to Top | Article Outline

MATERIALS AND METHODS

Data Source

We used data from the GWTG-R registry, which is an AHA sponsored, prospective, quality improvement registry of IHCA in the United States. The data collection process and reliability have been described in extensive detail previously (10–14). In the registry, cardiac arrest is defined as pulselessness, or a pulse with inadequate perfusion, requiring chest compressions and/or defibrillation, with a hospital-wide or unit-based emergency response by acute care facility personnel. Cases are identified, and data extracted by trained personnel from cardiac arrest flow sheets, hospital paging system logs, routine checks of code carts, pharmacy drug records, and hospital billing charges for resuscitation medication (15). The registry uses Utstein-style templates for cardiac arrest, standardized reporting guidelines used to define patient variables and outcomes, to facilitate uniform reporting across hospitals (16, 17). Hospital-level data were obtained from the American Hospital Association’s Annual Survey from 2013 (18).

Back to Top | Article Outline

Study Population

The cohort included events submitted to the GWTG-R registry between January 2000 and December 2016. We included all patients younger than 18 years old who received chest compressions while pulseless with a documented nonshockable initial rhythm (pulseless electrical activity or asystole) and who received at least one epinephrine bolus during the resuscitation. We included index events only from hospitals with at least 6 months of reporting and at least five pediatric cases with initial nonshockable rhythm reported. We excluded patients with the following: 1) cardiac arrest in the delivery room; 2) an illness category of trauma or hospital visitor; 3) vasopressor (epinephrine, norepinephrine, phenylephrine, and/or dopamine [for dopamine, at least 3 μg/kg/min]) infusion at the time of cardiac arrest; 4) treatment with extracorporeal membrane oxygenation during the event; 5) vasopressin received before epinephrine; 6) epinephrine given before loss of pulse; 7) epinephrine received after ROSC; 8) epinephrine given more than 20 minutes after loss of pulse; 9) missing data on covariates; 10) missing data on time to first epinephrine dose; and 11) missing data on in-hospital survival. Exclusion criteria were chosen based on prior work by Andersen et al (9). Our final cohort comprised 1,462 patients at 69 hospitals (Fig. 1).

Figure 1

Figure 1

Back to Top | Article Outline

Study Variables and Outcomes

The primary outcome measure was the relationship between the hospital rate of delayed epinephrine administration and survival to hospital discharge. Based on current guideline recommendations for administration of epinephrine (6) and prior work (9), delayed epinephrine administration was defined as more than 5 minutes between the time at which the need for compressions was identified and the administration of epinephrine. Secondary outcomes were the relationship between delayed epinephrine administration and ROSC (defined as at least 20 min with a palpable pulse), 24-hour survival, and survival with favorable neurologic outcome. Neurologic outcome was assessed in survivors with the use of the Pediatric Cerebral Performance Category (PCPC) score (19), where a PCPC score of 1 indicates no neurologic deficit, 2 mild cerebral disability, 3 moderate cerebral disability, 4 severe cerebral disability, and 5 coma or vegetative state. A PCPC score of 1–2 was considered a favorable neurologic outcome, and a PCPC score of 3–5 or death was considered a poor neurologic outcome. However, there is currently no universal definition of a favorable neurologic outcome in pediatric cardiac arrest patients using the PCPC score, and multiple definitions have been used previously (4, 11, 14).

Patient-level information included age, gender, and illness category. Cardiac arrest event characteristics included initial rhythm of asystole or pulseless electrical activity (PEA), location (ICU, telemetry, nonmonitored ward), time of day (daytime vs after hours), day of the week (weekday vs weekend), calendar year of admission, time to initiation of chest compressions, time to epinephrine administration, whether the event was witnessed or monitored, interventions in place at the time of cardiac arrest (e.g., mechanical ventilation), and whether an endotracheal tube was inserted during the event. Hospital-level characteristics included total number of hospital beds, proportion of ICU beds, geographic region, teaching status, type of hospital (primarily children or primarily adult), eligible arrests per 1,000 pediatric admissions, and number of years of GWTG-R participation. These 13 patient/event-level characteristics and the seven hospital-level characteristics are summarized in eTable 1 (Supplemental Digital Content 1, http://links.lww.com/PCC/A859).

Back to Top | Article Outline

Statistical Analyses

For each event, we calculated the time interval to epinephrine administration to be the difference between the time of first epinephrine administration and the time of cardiac arrest in minutes. We then calculated the hospital rates of delayed administration to be the number of hospital events with time to epinephrine administration greater than 5 minutes divided by the number of IHCA events reported by that hospital. We conducted univariate analyses of the associations between patient/event- and hospital-level characteristics and quartiles of hospital rates of delayed epinephrine administration using Pearson chi-square tests of significance for categorical variables and Kruskal-Wallis tests for continuous ones. We then constructed a multivariable hierarchical regression model adjusting for significant univariate associations (p < 0.05) of patient/event- and hospital-level characteristics with hospital rates of delayed epinephrine administration. In this multilevel model, we included hospital site as a random effect and adjusted for five patient/event-level and two hospital-level characteristics as fixed effects. We used the median odds ratio (OR) to quantify the extent to which hospital variation in rates of delayed administration of epinephrine can be explained by differences in hospital characteristics. In general, the median OR is obtained from a multivariable model that incorporates only patient/event-level variables, is always at least one, does not have a CI, and permits meaningful comparisons between aggregate hospital-level characteristics and patient/event-level variables. For example, a median OR of 1.5 suggests that the odds of delay in epinephrine administration for a patient are 50% higher at one randomly selected hospital than at another randomly selected hospital for a similar patient. After quantifying the extent to which differences across hospitals explain the variation in rates of delayed epinephrine administration, we added hospital characteristics to the previously selected patient/event-level variables to build our final multivariable regression model.

Last, we examined the extent to which variation in rates of delayed epinephrine administration across hospitals was associated with both patient/event-level survival and hospital-level survival. In the event-level models, we considered event-level ROSC, 24-hour survival, survival to discharge, and survival to discharge with favorable neurologic outcome to be dependent variables and included hospital rates of delayed epinephrine as an additional covariate in the previously described two-level hierarchical regression models. For the hospital-level analyses, we fit multivariable regression models in which we considered hospital rates of ROSC, 24-hour survival, survival to discharge, and survival to discharge with favorable neurologic outcome to be dependent variables and hospital bed size and hospital rates of delayed epinephrine to be independent variables (20–23).

For all analyses, the null hypothesis was evaluated at a two-sided significance level of .05 with 95% CIs. All analyses were conducted with SAS statistical software (Version 9.4; SAS Institute Inc, Cary, NC). The study was reviewed by the Medical City Dallas Institutional Review Board, which waived the requirement for informed consent.

Back to Top | Article Outline

Sensitivity Analysis

We conducted four sensitivity analyses to determine the robustness of our findings. First, to determine whether our findings were influenced by our definition of delayed epinephrine administration, we repeated the analysis of our primary outcome using a threshold of 3 minutes instead of 5 minutes to define delayed epinephrine administration. Second, given that delays in epinephrine administration and other aspects of resuscitation response may differ in patients who arrest in an ICU in comparison with patients who arrest outside an ICU, we repeated our primary analysis of hospital variation in epinephrine administration and its association with survival after restricting our cohort only to patients with the IHCA event in an ICU. Our third sensitivity analysis limited our cohort to patients with an IHCA event outside the ICU, and last, we repeated the primary analysis after restricting our cohort to include hospitals with greater than or equal to 10 eligible events instead of greater than or equal to five eligible events.

Back to Top | Article Outline

RESULTS

We identified 1,462 pediatric patients (> 18 yr) at 69 hospitals with an IHCA due to a nonshockable rhythm who were not receiving vasopressors at the time of the arrest and who received at least one dose of epinephrine between 2000 and 2016 in an eligible hospital (Fig. 1). Overall, 218 of 1,462 patients (14.9%) had a time to epinephrine administration of more than 5 minutes. The proportion of events with delayed epinephrine administration varied widely across hospitals and ranged from 0% to 80% (median, 15.6%; interquartile range, 7.1–25%) (Fig. 2). Tables 1 and 2 summarize patient and hospital characteristics across hospital quartiles of delayed epinephrine administration. Hospitals were categorized into quartiles based on the proportion of events with delayed epinephrine administration as follows: Q1, 0% to < 7.1%; Q2, 7.1% to < 15.6%; Q3, 15.6% to < 25%; and Q4, 25–80%.

TABLE 1

TABLE 1

TABLE 2

TABLE 2

Figure 2

Figure 2

In comparison with patients at hospitals in the lowest quartile of delay (Q1), patients at hospitals in the highest delay quartile (Q4) were less likely to be monitored (86.3% in Q1 vs 70.8% in Q4), have a witnessed event (94.0% in Q1 vs 74.0% in Q4), and have preexisting mechanical ventilation (51.2% in Q1 vs 29.2% in Q4). Hospitals in the highest quartile of epinephrine delay had higher eligible arrests per 1,000 admissions (mean ± SD, 0.53 ± 0.70 in Q1 vs 1.00 ± 2.12 in Q4, p = 0.0028) and had fewer events in the cohort (mean ± SD, 16.53 ± 13.01 in Q1 vs 7.38 ± 3.55 in Q4, p ≤0.0001). The median OR for delayed epinephrine administration after adjusting for differences in patient characteristics across hospitals was 1.23, suggesting that the odds of delay in epinephrine administration were 23% higher at one randomly selected hospital in comparison with a similar patient at another randomly selected hospital. On multivariable analysis, there were two patient/event-level predictors of delayed epinephrine administration: asystole as the first pulseless rhythm (OR, 1.54; 95% CI, 1.10–2.16; p = 0.0115) (Table 3) and insertion of an endotracheal tube during the event (OR, 1.86; 95% CI, 1.27–2.73; p = 0.0115) (Table 3). When the model was adjusted further for individual hospital-level characteristics, a smaller hospital size was found to be associated with higher rates of delayed epinephrine administration (< 200 vs ≥ 500 beds OR, 3.07 [95% CI, 1.22–7.73] p value for trend, 0.0175), whereas ICU arrest location was associated with lower rates of delayed epinephrine administration (OR, 0.51 [95% CI, 0.36–0.74] p value for trend, 0.0004) (Table 3).

TABLE 3

TABLE 3

The overall unadjusted patient-level rate of survival to hospital discharge was 33% (479/1462) and of those who survived to hospital discharge, 60% (286/479) had a favorable neurologic outcome. An additional 40% (193/479) survived to hospital discharge but without a documented PCPC score. ROSC occurred in 892 (61%) and survival at 24 hours after arrest occurred in 746 (51%) (Table 4). Hospital rate of delayed epinephrine was inversely correlated with unadjusted rates of ROSC (ρ = −0.54; p ≤ 0.001), 24-hour survival (ρ = −0.54; p ≤ 0.001), and survival to hospital discharge (ρ = −0.39; p = 0.001), but not survival to discharge with favorable neurologic outcome (ρ = −0.15; p = 0.222) (Fig. 3). Unadjusted hospital rates of delayed epinephrine were significantly associated with worse hospital rates of ROSC (p = 0.003) and survival at 24 hours (p < 0.001), but not with survival to discharge or survival to discharge with favorable neurologic outcome (p = 0.11 in both cases) (Table 5). These relationships held after adjustment for hospital bed size. After hierarchical regression model adjustment for patient- and hospital-level characteristics, higher hospital epinephrine delay rates were significantly associated with worse patient-level ROSC (p = 0.019) and survival at 24 hours (p = 0.018), but not with survival to discharge or survival to discharge with favorable neurologic outcome (p = 0.20 and p = 0.16, respectively).

TABLE 4

TABLE 4

TABLE 5

TABLE 5

Figure 3

Figure 3

The results of our sensitivity analyses closely aligned with those of our primary analysis. We redefined delayed epinephrine administration using a cutoff of greater than 3 minutes rather than greater than 5 minutes after onset of cardiac arrest. In these analyses, delayed epinephrine administration remained inversely correlated with ROSC (ρ = −0.52, p < 0.001) and 24-hour survival (ρ = −0.51; p < 0.001) but was also inversely correlated with survival to discharge (ρ = −0.43; p < 0.001). After restricting the CPA events to an ICU location only, the number of events was reduced from 1,462 to 917 and the number of hospitals reduced to 48. In this subcohort, the proportion of events with delayed epinephrine administration ranged from 0% to 40%, suggesting less variability than in the main study sample. Hierarchical regression analyses for this subcohort found no association with delayed epinephrine administration and outcomes, noting that patient-level survival outcomes do not vary as much when the analysis is restricted to the ICU. Restricting CPA events to those occurring outside the ICU yielded a subcohort of 395 events and 31 hospitals. After hierarchical regression model adjustment for patient- and hospital-level characteristics, higher hospital epinephrine delay rates were significantly associated with worse patient-level survival at 24 hours (p = 0.0028) and survival to discharge (p = 0.035), but not with ROSC (p = 0.20). We were not able to fit a hierarchical regression model for the survival to discharge with favorable neurologic outcome because of the limited sample size. Finally, the cohort was limited to hospitals with greater than or equal to 10 cases rather than greater than or equal to 5 cases of nonshockable IHCA, which reduced the number of events from 1,462 to 1,278 and the number of hospitals from 69 to 40. These analyses noted that at the hospital level delayed epinephrine administration was associated with worse 24-hour survival and survival to hospital discharge. Throughout all analyses, hospitals in the “worst” quartile with respect to delayed administration of epinephrine were associated with much lower survival rates, both at the patient and hospital levels, than other hospitals.

Back to Top | Article Outline

DISCUSSION

In a large, multicenter registry of pediatric IHCAs, we found large variation in the timely administration of epinephrine for patients with an initial nonshockable rhythm of pulseless electric activity or asystole, with overall hospital rates of delayed epinephrine ranging from 0% to 80%. The observed hospital-level differences in delayed epinephrine administration can only partly be explained by certain patient and cardiac arrest event characteristics. Importantly, hospital-level differences explain a certain amount of this variation; however, many of the individual hospital characteristics that we explored—such as cardiac arrest volume, academic status, and type of hospital—were unrelated to hospital performance in delayed epinephrine administration. In multivariable analyses, there were two patient/event-level predictors of delayed epinephrine administration, asystole, and insertion of an endotracheal tube. Among traditional hospital factors evaluated, only bed size less than 200 beds compared to greater than or equal to 500 beds and ICU location were associated with differences in rates of delayed epinephrine administration. We also found that hospital rates of ROSC, 24-hour survival, and survival to discharge were lower at hospitals with more frequent delays in epinephrine administration than in hospitals where such delays were less common. Throughout all analyses, hospitals in the “worst” quartile with respect to delayed administration of epinephrine were associated with much lower survival rates, both at the patient and hospital-level, than other hospitals.

In recent adult (8) and pediatric (9) patient-level analysis, delay in administration of epinephrine for IHCA with an initial nonshockable rhythm was associated with worse outcomes. Patients with a nonshockable initial rhythm and prompt administration of epinephrine had improved ROSC, short-term survival, survival to hospital discharge, and survival to hospital discharge with a promising neurologic outcome as compared to patients who had a delay in initial epinephrine dose. These associations remained when accounting for multiple predetermined potentially confounding patient, event, and hospital characteristics and in multiple different sensitivity analyses in both studies (8, 9). Our study adds to this growing literature for pediatric IHCA by showing marked hospital-level variation in delays to epinephrine administration for nonshockable pediatric IHCA. Hospitals with lower rates of delay to epinephrine administration had higher rates of ROSC, 24-hour survival, and survival to hospital discharge than those hospitals with higher rates of delayed epinephrine administration.

The physiologic rationale for early administration of epinephrine in patients with cardiac arrest is strong, particularly in those with rhythms not amenable to defibrillation. Epinephrine is a potent peripheral vasoconstrictor as well as a coronary artery vasodilator. This combination of physiologic effects results in an increase in coronary perfusion pressure, which has been shown to be strongly associated with ROSC in both animals and humans (24, 25). Epinephrine is currently recommended in pediatric cardiac arrests as the first-line pharmacologic intervention despite no randomized placebo-controlled trials in this patient population (26, 27). One randomized placebo-controlled study in the adult out-of-hospital cardiac arrest population found improved ROSC and short-term survival with administration of epinephrine (28). However, the study was underpowered to detect any difference in long-term outcome because of unanticipated lack of enrollment (29).

Unlike adults, pediatric CPA is more likely to be secondary to respiratory failure and shock, rather than a primary cardiac etiology, with the most common presenting rhythm being PEA or asystole. Yet, overall survival outcomes in children after IHCA are superior to adults despite fewer shockable rhythms (3). The prospective trial by Tibballs and Kinney (5) demonstrated that 90% of pediatric IHCA had an initial cardiac rhythm of PEA, asystole, or bradycardia. Studies have also demonstrated that use of epinephrine during resuscitation has significant effect on rhythm changes during resuscitation. In patients with PEA as an initial rhythm, epinephrine can improve ROSC but increases the rate of rhythm transitions, including transient ROSC, during resuscitation and may render the patient more unstable overall (30).

Current guidelines recommend administration of epinephrine as soon as vascular or intraosseous access is obtained and subsequently every 3–5 minutes for patients with a nonshockable rhythm (6). Therefore, the variation in rate of delayed epinephrine administration observed in our study was particularly striking, with a median OR of 1.23. Although higher volume is associated with improved outcomes for a number of surgical and medical procedures, we did not find a relationship between IHCA volume and delayed time to epinephrine administration (31–34). Thus, we cannot conclude that a lack of institutional experience at low-volume pediatric resuscitation sites is an important underlying factor. This is in contrast to an adult GWTG-R study evaluating hospital variation in delayed epinephrine administration in adult nonshockable IHCA, which found delays in epinephrine administration were more common at hospitals with low cardiac arrest case volume than in high-volume hospitals (35). Thus, we may not point toward a lack of institutional experience at low-volume pediatric sites with regard to resuscitation care as an important underlying factor. We did not find an association between hospital structural factors such as teaching status or type of hospital (primarily children or primarily adult) with delays in epinephrine administration. This is similar to a previous study showing substantial hospital variation in rates of delayed defibrillation for adult IHCA attributable to shockable rhythm (36). Similar to our study, delays in defibrillation were not associated with hospital structural characteristics, with the exception of bed size and ICU location, as was noted in our study. The paucity of conventional hospital-level characteristics associated with performance of epinephrine administration time may reflect a true opportunity for future improvement because it suggests the importance of process of care measures in achieving best practices. These findings suggest that hospital processes of care may be more important in determining hospital quality. Indeed, many of the characteristics we evaluated are not even readily amenable to modification (e.g., admission volume, academic status, geographic region, and type of hospital). The association between bed volume and epinephrine administration time seen in this study may actually reflect the impact of quality improvement efforts or even active intervention trials to improve resuscitation outcomes at larger hospitals rather than the hospital size itself.

As noted above, when the multivariable model was adjusted for individual hospital-level characteristics, an ICU arrest location was associated with lower rates of delayed epinephrine administration. Additionally, on sensitivity analysis after restricting the CPA events to an ICU location only, there was less variability in epinephrine delay rates than in the main study sample, and hierarchical regression analyses found no association with delayed epinephrine administration and outcomes in this cohort. However, when we looked at events occurring outside the ICU, the hierarchical regression model noted that higher hospital epinephrine delay rates were significantly associated with worse patient-level survival at 24 hours and survival to discharge. As noted in the outcomes table and the correlations figure, it is the events in the fourth quartile or highest rate of epinephrine delay that are vastly different from the events in the other three quartiles of delay. Looking at differences in characteristics in this fourth quartile, the higher rates of tracheal intubation may relate to lower rates of mechanical ventilation at the time of arrest (i.e., you cannot intubate the de novo arresting child who is already intubated on a mechanical ventilator). Children on mechanical ventilation must also be in the ICU, so not surprisingly the fourth quartile has the least number of children in the ICU at the time of IHCA. Hence, this “worst outcome” quartile consists of a large group of children not in the ICU, and not on mechanical ventilation with the lowest rate of monitoring and witnessed arrest. Thus, is the delay in epinephrine merely a marker for not being in the ICU, not being witnessed/monitored at the time of the IHCA, and as a result having a less prepared/educated/trained staff where epinephrine dosing is forgotten? Another factor to consider is that the majority of these hospitals are primary adult hospitals, and are these children being kept on the floor at these centers who would normally be transferred to an ICU or monitored area in a pediatric focused hospital? This may be easily addressed with the increasing use of electronic medical record-based systems that identify higher acuity and/or deteriorating patients and in some cases automatically activating a rapid response or medical emergency team. The failure to recognize the deteriorating child could be remedied in an automated fashion and could initiate transfer of the child to a dedicated children’s hospital if no suitable PICU/step down bed is available at that hospital.

Back to Top | Article Outline

LIMITATIONS

A number of limitations should be considered when interpreting the current study. First, the data are observational, and the possibility of unmeasured confounding remains. We tried to account for this by multivariable regression modeling, including adjusting for patient, event, and hospital-level characteristics. Second, we excluded a small number of patients based on missing values for covariates, time to epinephrine, or the outcomes, which might decrease the generalizability of our results. Third, GWTG-R is a quality improvement registry, and hospital participation is voluntary. Given the fact that a minority of children’s hospitals in the United States participate in the registry and hospitals in a quality improvement registry are more likely to direct substantial resources to improving CPR outcomes, our findings may not be generalizable to nonregistry hospitals. Fourth, although our model adjusted for several patient, event and hospital characteristics, we did not have information on certain factors. For instance, we did not have information on hospital factors such as staffing ratios, presence of around-the-clock intensivists in critical care units, and use of mock codes and other quality improvement initiatives. It is possible that delays in epinephrine administration are confounded by other components of CPR efforts, including quality of chest compressions, which is difficult to quantify and is not measured in GWTG-R. Fifth, functional status has been inferred from PCPC scores at discharge. Although a favorable PCPC score at discharge is associated with improved long-term survival (19), patient performance on dedicated scales for neurologic assessment were not available, and we were therefore only able to assess functional status. Additionally, a PCPC score was only reported in 60% of survivors. Finally, the cohort was restricted to a specific pediatric patient population which consisted of 1,462 patients among 19,957 pediatric IHCA events (7.3%) and did not include those patients receiving vasopressors at the time of the arrest, thus these results may not be generalizable to different pediatric patient populations who suffer IHCA.

Back to Top | Article Outline

CONCLUSIONS

Although delays in epinephrine administration following pediatric IHCA with an initial nonshockable rhythm of pulseless electric activity or asystole are common, there is substantial hospital variation in rates of delayed epinephrine administration. Differences across hospitals explained a substantial degree of the variation in rates of delayed epinephrine administration, but few facility characteristics were found to explain this variation. Asystole and insertion of an endotracheal tube were the only patient/event-level predictors of delayed epinephrine administration, and bed volume less than 200 beds compared with greater than or equal to 500 beds and ICU location were the only hospital-level characteristics found to be associated with delayed epinephrine administration. Hospital rates of ROSC, 24-hour survival, and survival to discharge are inversely correlated with hospital rates of delayed epinephrine. After adjusting for relevant factors, patient- and hospital-level delayed epinephrine administration was associated with lower event and 24-hour survival across increasing quartiles of epinephrine delay. Given extensive differences in epinephrine administration time across institutions and the recognized impact of delayed epinephrine on survival, new approaches to improve hospital performance in epinephrine administration time could represent a critical area for quality improvement. Further studies are needed to determine if improving hospital performance on time to epinephrine administration, especially at hospitals with poor performance on this metric, will lead to improvement in outcomes.

Back to Top | Article Outline

ACKNOWLEDGMENTS

Get With the Guidelines-Resuscitation Investigators: Besides the authors Tia T. Raymond, MD, Vinay M. Nadkarni MD, and Chris S. Parshuram MB ChB, PhD, members of the Get With the Guidelines-Resuscitation Pediatric Research Task Force include: Anne-Marie Guerguerian, Robert M. Sutton, Melania Bembea, Dianne L. Atkins, Elizabeth Foglia, Ericka Fink, Javier J. Lasa, Joan Roberts, Jordan Duval-Arnould, Michael Gaies, Monica Kleinman, Punkaj Gupta, and Taylor Sawyer.

Back to Top | Article Outline

REFERENCES

1. Topjian AA, Berg RA, Nadkarni VM. Pediatric cardiopulmonary resuscitation: Advances in science, techniques, and outcomes. Pediatrics 2008; 122:1086–1098
2. Berg MD, Nadkarni VM, Zuercher M, et al. In-hospital pediatric cardiac arrest. Pediatr Clin North Am 2008; 55:589–604, x
3. Nadkarni VM, Larkin GL, Peberdy MA, et al; National Registry of Cardiopulmonary Resuscitation Investigators: First documented rhythm and clinical outcome from in-hospital cardiac arrest among children and adults. JAMA 2006; 295:50–57
4. Girotra S, Spertus JA, Li Y, et al; American Heart Association Get With the Guidelines–Resuscitation Investigators: Survival trends in pediatric in-hospital cardiac arrests: An analysis from Get With the Guidelines-Resuscitation. Circ Cardiovasc Qual Outcomes 2013; 6:42–49
5. Tibballs J, Kinney S. A prospective study of outcome of in-patient paediatric cardiopulmonary arrest. Resuscitation 2006; 71:310–318
6. de Caen AR, Berg MD, Chameides L, et al. Part 12: pediatric advanced life support: 2015 American Heart Association Guidelines update for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation 2015; 132:S526–S542
7. Atkins DL, Berger S, Duff JP, et al. Part 11: pediatric basic life support and cardiopulmonary resuscitation quality: 2015 American Heart Association Guidelines Update for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation 2015; 132:S519–S525
8. Donnino MW, Salciccioli JD, Howell MD, et al; American Heart Association’s Get With The Guidelines-Resuscitation Investigators: Time to administration of epinephrine and outcome after in-hospital cardiac arrest with non-shockable rhythms: Retrospective analysis of large in-hospital data registry. BMJ 2014; 348:g3028
9. Andersen LW, Berg KM, Saindon BZ, et al; American Heart Association Get With the Guidelines–Resuscitation Investigators: Time to epinephrine and survival after pediatric in-hospital cardiac arrest. JAMA 2015; 314:802–810
10. Peberdy MA, Ornato JP, Larkin GL, et al; National Registry of Cardiopulmonary Resuscitation Investigators: Survival from in-hospital cardiac arrest during nights and weekends. JAMA 2008; 299:785–792
11. Matos RI, Watson RS, Nadkarni VM, et al; American Heart Association’s Get With The Guidelines–Resuscitation (Formerly the National Registry of Cardiopulmonary Resuscitation) Investigators: Duration of cardiopulmonary resuscitation and illness category impact survival and neurologic outcomes for in-hospital pediatric cardiac arrests. Circulation 2013; 127:442–451
12. Meaney PA, Nadkarni VM, Cook EF, et al; American Heart Association National Registry of Cardiopulmonary Resuscitation Investigators: Higher survival rates among younger patients after pediatric intensive care unit cardiac arrests. Pediatrics 2006; 118:2424–2433
13. Donoghue AJ, Nadkarni VM, Elliott M, et al; American Heart Assocation National Registry of Cardiopulmonary Resuscitation Investigators: Effect of hospital characteristics on outcomes from pediatric cardiopulmonary resuscitation: A report from the national registry of cardiopulmonary resuscitation. Pediatrics 2006; 118:995–1001
14. Ortmann L, Prodhan P, Gossett J, et al; American Heart Association’s Get With the Guidelines–Resuscitation Investigators: Outcomes after in-hospital cardiac arrest in children with cardiac disease: A report from Get With the Guidelines–Resuscitation. Circulation 2011; 124:2329–2337
15. Peberdy MA, Kaye W, Ornato JP, et al. Cardiopulmonary resuscitation of adults in the hospital: A report of 14720 cardiac arrests from the National Registry of Cardiopulmonary Resuscitation. Resuscitation 2003; 58:297–308
16. Cummins RO, Chamberlain D, Hazinski MF, et al. Recommended guidelines for reviewing, reporting, and conducting research on in-hospital resuscitation: The in-hospital ‘Utstein style’. American Heart Association. Circulation 1997; 95:2213–2239
17. Jacobs I, Nadkarni V, Bahr J, et al; International Liaison Committee on Resuscitation; American Heart Association; European Resuscitation Council; Australian Resuscitation Council; New Zealand Resuscitation Council; Heart and Stroke Foundation of Canada; InterAmerican Heart Foundation; Resuscitation Councils of Southern Africa; ILCOR Task Force on Cardiac Arrest and Cardiopulmonary Resuscitation Outcomes: Cardiac arrest and cardiopulmonary resuscitation outcome reports: Update and simplification of the Utstein templates for resuscitation registries: A statement for healthcare professionals from a task force of the International Liaison Committee on Resuscitation (American Heart Association, European Resuscitation Council, Australian Resuscitation Council, New Zealand Resuscitation Council, Heart and Stroke Foundation of Canada, InterAmerican Heart Foundation, Resuscitation Councils of Southern Africa). Circulation 2004; 110:3385–3397
18. The American Hospital Association: AHA Annual Survey Database Fiscal Year 2013. 2014. Available at: http://ahadata.com. Accessed December 27, 2018
19. Fiser DH. Assessing the outcome of pediatric intensive care. J Pediatr 1992; 121:68–74
20. Goldstein H. Multilevel Statistical Models. 1995London, England, Edward Arnold.
    21. Larsen K, Petersen JH, Budtz-Jørgensen E, et al. Interpreting parameters in the logistic regression model with random effects. Biometrics 2000; 56:909–914
      22. Larsen K, Merlo J. Appropriate assessment of neighborhood effects on individual health: Integrating random and fixed effects in multilevel logistic regression. Am J Epidemiol 2005; 161:81–88
        23. Goldstein H, Browne W, Rasbash J. Partitioning variation in multilevel models. Underst Stat 2002; 1:223–232
          24. Paradis NA, Martin GB, Rosenberg J, et al. The effect of standard- and high-dose epinephrine on coronary perfusion pressure during prolonged cardiopulmonary resuscitation. JAMA 1991; 265:1139–1144
          25. Paradis NA, Martin GB, Rivers EP, et al. Coronary perfusion pressure and the return of spontaneous circulation in human cardiopulmonary resuscitation. JAMA 1990; 263:1106–1113
          26. Kleinman ME, Chameides L, Schexnayder SM, et al. Part 14: Pediatric advanced life support: 2010 American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation 2010; 122:S876–S908
          27. Bingham R, Eich C, et al. European Resuscitation Council guidelines for resuscitation 2010 section 6: Paediatric life support. Resuscitation 2010; 81:1364–1388
          28. Jacobs IG, Finn JC, Jelinek GA, et al. Effect of adrenaline on survival in out-of-hospital cardiac arrest: a randomised double-blind placebo-controlled trial. Resuscitation 2011; 82:1138–1143
          29. Olasveengen TM. Adrenaline for out of hospital cardiac arrest? BMJ 2013; 347:f7268
          30. Nordseth T, Olasveengen TM, Kvaløy JT, et al. Dynamic effects of adrenaline (epinephrine) in out-of-hospital cardiac arrest with initial pulseless electrical activity (PEA). Resuscitation 2012; 83:946–952
          31. Birkmeyer JD, Stukel TA, Siewers AE, et al. Surgeon volume and operative mortality in the United States. N Engl J Med 2003; 349:2117–2127
          32. Dudley RA, Johansen KL, Brand R, et al. Selective referral to high-volume hospitals: Estimating potentially avoidable deaths. JAMA 2000; 283:1159–1166
          33. Canto JG, Every NR, Magid DJ, et al. The volume of primary angioplasty procedures and survival after acute myocardial infarction. National Registry of Myocardial Infarction 2 Investigators. N Engl J Med 2000; 342:1573–1580
          34. Halm EA, Lee C, Chassin MR. Is volume related to outcome in health care? A systematic review and methodologic critique of the literature. Ann Intern Med 2002; 137:511–520
          35. Khera R, Chan PS, Donnino M, et al; American Heart Association’s Get With The Guidelines-Resuscitation Investigators: Hospital variation in time to epinephrine for nonshockable in-hospital cardiac arrest. Circulation 2016; 134:2105–2114
          36. Chan PS, Nichol G, Krumholz HM, et al; American Heart Association National Registry of Cardiopulmonary Resuscitation (NRCPR) Investigators: Hospital variation in time to defibrillation after in-hospital cardiac arrest. Arch Intern Med 2009; 169:1265–1273
          Keywords:

          adrenaline; epinephrine; hospital variation; in-hospital cardiac arrest; pediatric; survival

          Supplemental Digital Content

          Back to Top | Article Outline
          ©2019The Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies