Abstract
Aims:
To describe trends in pediatric in-hospital cardiac arrest drug administration and to assess temporal associations of the Pediatric Advanced Life Support (PALS) guideline changes with drug usage.
Methods:
Pediatric patients <18 years old with in-hospital cardiac arrest recorded in the American Heart Association Get With The Guidelines–Resuscitation database between 2002 and 2018 were included. The annual adjusted odds of receiving each intra-arrest medication was determined. The association between changes in the PALS Guidelines and medication use over time was assessed interrupted time series analyses.
Results:
A total of 6107 patients were analyzed. The adjusted odds of receiving lidocaine (0.33; 95% CI, 0.18, 0.61; p<0.001), atropine (0.19; 95% CI 0.12, 0.30; p<0.001) and bicarbonate (0.54; 95% CI 0.35, 0.86; p=0.009) were lower in 2018 compared to 2002. For lidocaine, there were no significant changes in the step (−2.1%; 95% CI, −5.9%, 1.6%; p=0.27) after the 2010 or 2015 (Step: −1.5%; 95% CI, −8.0%, 5.0; p=0.65) guideline releases. There were no significant changes in the step for bicarbonate (−2.3%; 95% CI, −7.6%, 3.0%; p = 0.39) after the 2010 updates. For atropine, there was a downward step change after the 2010 guideline release (−5.9%; 95% CI, −10.5%, −1.3%; p=0.01).
Conclusions:
Changes to the PALS guidelines for lidocaine and bicarbonate were not temporally associated with acute changes in the use of these medications; however, better alignment with these updates was observed over time. A minor update to the language surrounding atropine in the PALS text was associated with a modest acute change in the observed use of atropine. Future studies exploring other factors that influence prescribers in pediatric IHCA are needed.
Keywords: Pediatric Advanced Life Support, guidelines, cardiac arrest
INTRODUCTION:
The American Heart Association’s (AHA) Pediatric Advanced Life Support (PALS) guidelines1–4 provide recommendations to guide resuscitative efforts for pediatric cardiac arrest. Since the original guidelines were released, updates to the PALS cardiac arrest algorithms have included modifications to the text and pictorial flowcharts related to intra-arrest drug administration. Notably, adult data have shown rapid adaptation of intra-arrest medication practices for some medications following guideline changes in Advanced Cardiac Life Support (ACLS).5–7 Despite key changes in the pediatric cardiac arrest guidelines between 2002 and 2018, a similar assessment of intra-arrest medication use has not yet been explored. These data may identify practices which appear to diverge from the guidelines, providing important feedback on areas which may need more emphasis during provider education and instructor training sessions and ultimately lead to improved outcomes.8–10
In the present study, we leveraged the AHA Get With The Guidelines - Resuscitation (GWTG-R) pediatric database to describe trends in pediatric in-hospital cardiac arrest (IHCA) drug administration. We additionally examined temporal associations of the AHA PALS guidelines with medication usage. We hypothesized that trends in use of intra-arrest drugs would increase or decrease in the years following a change in the guidelines that strengthened or weakened a specific medication recommendation, respectively.
METHODS:
Data Source
The GWTG-R pediatric database is a large, prospectively collected, quality-improvement registry of IHCA events sponsored by the AHA. Hospitals participating in the registry submit clinical information regarding the medical history, peri-resuscitation care, and outcomes of consecutive patients who experienced IHCA using an online, interactive case report form and Patient Management Tool™ (IQVIA, Parsippany, New Jersey). The design, data collection, and quality control of the repository have been described previously.11, 12 Data are collected prospectively on all IHCA patients in participating hospitals who do not have prior do-not-resuscitate orders. Hospital-level data were obtained from the 2018 American Hospital Association Annual Survey.13 All participating institutions were required to comply with local regulatory and privacy guidelines. Because data were used primarily at the local site for quality improvement, sites were granted a waiver of informed consent under the common rule. Additionally, this research was classified as nonhuman research by the Boston Children’s Hospital Institutional Review Board. IQVIA is the data collection coordination center for the American Heart Association/American Stroke Association Get With The Guidelines® programs.
Study Population
Index IHCA events in the GWTG-R pediatric database in patients 17 years or younger between January 1, 2002 and December 31, 2018 were included. Events from the years 2000 and 2001 (the first 2 years of the registry), events listed as occurring in an outpatient, rehabilitation, adult care, delivery room, nursery or neonatal intensive care area, events related to hospital visitors or employees, events missing data for any of the included covariates (see below) and events with no chest compressions or compressions for less than one minute were excluded. Non-pulseless events (chest compressions administered for bradycardia with poor perfusion) were also excluded given the differences in medication recommendations for this condition.
Medication Exposures and Definitions
The GWTG-R database includes medications provided following pulselessness and prior to return of spontaneous circulation (ROSC) or death. These medications include epinephrine, vasopressin, amiodarone, lidocaine, calcium, magnesium, atropine, dextrose, and bicarbonate. Information on dose is not provided for any medications and the timing of administration is collected for epinephrine only. Shockable rhythm is defined as pulseless ventricular tachycardia (VT) or ventricular fibrillation (VF), and non-shockable rhythm is defined as pulseless electrical activity (PEA) or asystole.
Updates to PALS Guidelines
We hypothesized that medication prescribing patterns would correlate with changes to the PALS algorithm in the years following guideline updates. PALS guidelines documents from 2000, 2005, 2010 and 20151–4 were screened for the medications included in the GWTG- R registry. We collected the exact wording of the recommendation for each medication and noted presence or absence of the medication in the pictorial algorithm (when relevant) for each release year prior to analysis (Supplementary Table 1). We a priori planned to assess changes in lidocaine use before and after 2010 and 2015 and bicarbonate use before and after 2010. These time points and medications were selected to reflect major changes in the pictorial algorithm or language surrounding the recommendation. We also performed post hoc analyses for atropine before and after 2010 to reflect a minor change in the PALS guideline text which coincided with a major change in the adult ACLS guidelines.
Statistical Analysis
Descriptive statistics were used to characterize the study population. Continuous variables are reported as medians with interquartile ranges (IQRs). Categorical variables are reported as counts with frequencies. Frequency of use for each medication were calculated for each year and presented graphically over time.
For each medication of interest, multivariable logistic regression with generalized estimating equations (GEEs) and an exchangeable variance-covariance structure was used to assess the relationship between event year and medication. For these analyses, year was considered as a categorical variable (2002 used as a reference). Covariates in the GEE model, chosen by the authors as those most likely to be associated with prescribing patterns, including age, event location, initial rhythm, academic status of hospital, witnessed events and monitored events.
To assess the relationship between guideline updates and usage of lidocaine, sodium bicarbonate and atropine, interrupted time series analyses using segmented linear regression with GEE for each medication was performed. For large datasets linear regression is valid even for binary outcomes and often provides more easily interpretable results.14 The model included age, event location, initial rhythm, academic status of hospital, witnessed events and monitored events as covariables in addition to variables estimating the preintervention trend, step change following the intervention, and the postintervention trend. To reflect the timing of the updated guideline publication in October 2010 and November 2015, the preintervention segment included the entire years of 2010 and 2015, respectively. The step change refers to the immediate change in the intercept before and after the guideline release year. The slope change refers to the change in trend. Positive values signify an increase in medication use while negative values imply a decrease.
Sensitivity analyses were performed for lidocaine limited to events with initial shockable rhythm and events in which at least 1 antiarrhythmic medication (lidocaine, amiodarone or both) was administered. Additionally, to reflect the delay from the time from publication of the PALS guidelines to the publication of updates in the PALS Provider Manual, a post hoc sensitivity analysis was performed with the year 2011 included in the preintervention segment for the medications of interest and the subgroup analyses. To assess the effect of age on use of atropine, a post hoc subgroup analysis was performed according to age category at the time of the event defined as: Neonate (<1 month), Infant (1 month to <1 year), Child (1 year to <12 years) and Adolescent (≥12 years). Finally, to account for arrest duration, a post hoc subgroup analysis limited to prolonged events, defined as >10 minutes, was performed to assess the adjusted odds ratio of medication use over the study period. P values less than 0.05 were considered statistically significant. All statistics were performed using STATA, Version 14.2 (StataCorp LP, College Station, TX).
RESULTS:
Patient and Event Characteristics
A total of 6107 events reported by 311 hospitals were included (Supplementary Figure 1). Patient and event characteristics are shown in Table 1. The median age was 1.6 (0.3, 9) years and 44% were female. The majority of events had initial non-shockable rhythm (86%) and occurred within a pediatric intensive care unit (ICU; 66%). The duration from event recognition to ROSC or termination of resuscitation decreased over time from a median of 23 minutes (IQR: 12, 41 minutes) in 2002 to 14 minutes (IQR: 4, 34 minutes) in 2018 (p < 0.001; Supplementary Figure 2). Missing data for covariates are shown in Supplementary Table 2.
Table 1:
Patient, event and hospital characteristics.
| 2002 – 2005 n = 1175 |
2006 – 2010 n = 1689 |
2011 – 2015 n = 1749 |
2016 – 2018 n = 1494 |
|
|---|---|---|---|---|
| Demographics | ||||
| Sex* 1 missing | ||||
| Female | 506 (43) | 730 (43) | 783 (45) | 675 (45) |
| Male | 669 (57) | 959 (57) | 965 (55) | 819 (55) |
| Age group | ||||
| Neonate (<1 month) | 173 (15) | 311 (18) | 332 (19) | 121 (8) |
| Infant (1 month to <1 year) | 313 (27) | 468 (28) | 340 (19) | 442 (30) |
| Child (1 year to <12 years) | 396 (34) | 621 (37) | 730 (42) | 651 (44) |
| Adolescent (≥12 years) | 293 (25) | 289 (17) | 347 (20) | 280 (19) |
| Illness category | ||||
| Medical | ||||
| Cardiac | 185 (16) | 282 (17) | 292 (17) | 291 (19) |
| Non-cardiac | 503 (43) | 676 (40) | 781 (45) | 703 (47) |
| Surgical | ||||
| Cardiac | 177 (15) | 368 (22) | 375 (21) | 282 (19) |
| Non-cardiac* | 245 (21) | 306 (18) | 269 (15) | 218 (15) |
| Newborn† | 65 (6) | 57 (3) | 32 (2) | 0 (0) |
| Pre-existing conditions‡ *12 missing for all PMH | ||||
| Acute non-stroke CNS event | 168 (14) | 127 (8) | 114 (7) | 107 (7) |
| Baseline depression in CNS | 234 (20) | 279 (17) | 231 (13) | 251 (17) |
| Heart failure prior admission | 167 (14) | 108 (6) | 92 (5) | 86 (6) |
| Heart failure this admission | 186 (16) | 129 (8) | 109 (6) | 112 (8) |
| Hepatic insufficiency | 58 (5) | 68 (4) | 74 (4) | 76 (5) |
| Hypotension | 510 (43) | 624 (37) | 428 (25) | 397 (27) |
| Metabolic abnormalities | 259 (22) | 293 (17) | 238 (14) | 340 (23) |
| Metastatic malignancy | 60 (5) | 90 (5) | 113 (7) | 105 (7) |
| Pneumonia | 118 (10) | 136 (8) | 117 (7) | 116 (8) |
| Respiratory insufficiency | 711 (61) | 999 (59) | 925 (53) | 975 (65) |
| Renal insufficiency | 130 (11) | 152 (9) | 187 (11) | 187 (13) |
| Septicemia | 190 (16) | 241 (14) | 227 (13) | 171 (11) |
| Location and Time of Event | ||||
| Location of event | ||||
| Emergency department | 197 (17) | 276 (16) | 233 (13) | 167 (11) |
| Intensive care unit | 794 (68) | 1053 (62) | 1118 (64) | 1094 (73) |
| Floor | ||||
| Without telemetry | 84 (7) | 103 (6) | 102 (6) | 86 (6) |
| With telemetry | 19 (2) | 54 (3) | 29 (2) | 15 (1) |
| Procedural area | 61 (5) | 148 (9) | 185 (11) | 116 (8) |
| Other§ | 20 (2) | 55 (3) | 82 (5) | 16 (1) |
| Time of week | ||||
| Weekend|| | 362 (31) | 493 (29) | 492 (28) | 417 (28) |
| Weekday | 813 (69) | 1196 (71) | 1257 (72) | 1077 (72) |
| Time of day | ||||
| Nighttime# | 368 (32) | 471 (28) | 480 (28) | 459 (31) |
| Day time | 795 (68) | 1209 (72) | 1260 (72) | 1025 (69) |
| Event Characteristics | ||||
| Witnessed | ||||
| Yes | 1075 (91) | 1569 (93) | 1652 (94) | 1400 (94) |
| No | 100 (9) | 120 (7) | 97 (6) | 94 (6) |
| Monitored | ||||
| Yes | 1000 (85) | 1497 (89) | 1513 (87) | 1354 (91) |
| No | 175 (15) | 192 (11) | 236 (13) | 140 (9) |
| Pulseless rhythm | ||||
| Beginning of event | 901 (77) | 1291 (76) | 1252 (72) | 1082 (72) |
| During event | 274 (23) | 398 (24) | 497 (28) | 412 (28) |
| Initial pulseless rhythm | ||||
| Non-shockable | 975 (83) | 1473 (87) | 1505 (86) | 1313 (88) |
| Subsequent Shockable (% of initial non-shockable) | 131/975 (13) | 151/1473 (10) | 133/1505 (9) | 111/1313 (8) |
| Shockable | 200 (17) | 216 (13) | 244 (14) | 181 (12) |
| Hospital Characteristics | ||||
| Geographic region | ||||
| Northeast | 244 (21) | 199 (12) | 383 (22) | 274 (18) |
| South Atlantic | 283 (24) | 365 (22) | 497 (28) | 278 (19) |
| Midwest | 176 (15) | 261 (15) | 302 (17) | 434 (29) |
| South Central | 248 (21) | 581 (34) | 425 (24) | 262 (18) |
| West | 223 (19) | 283 (17) | 142 (8) | 246 (16) |
| Ownership *missing 278 | ||||
| Government/Military | 106 (10) | 44 (3) | 80 (5) | 32 (2) |
| Non-profit | 950 (87) | 1468 (94) | 1582 (93) | 1373 (94) |
| Private | 41 (4) | 53 (3) | 41 (2) | 59 (4) |
| Teaching status | ||||
| Non-teaching | 57 (5) | 42 (2) | 26 (1) | 16 (1) |
| Minor | 399 (34) | 400 (24) | 317 (18) | 433 (29) |
| Major | 719 (61) | 1247 (74) | 1406 (80) | 1045 (70) |
CNS denotes central nervous system, GWTG-R denotes Get With The Guidelines-Resuscitation.
Including non-cardiac surgical patients, trauma patients, and obstetric patients.
The newborn illness category was added to the GWTG-R registry in 2005 and removed in 2015.
Definitions have been provided elsewhere.[12]
Including ambulatory or outpatient clinics, diagnostic or interventional areas, operating room, post-anesthesia recovery room, rehabilitation unit, same-day surgical area, and delivery room.
Friday 11 PM to Monday 7 AM.
11:00 PM to 6:59 AM.
Lidocaine
Overall, lidocaine was used in 7.6% (95% CI, 6.9%, 8.3%) of arrests including 29.6% (26.5%, 32.8%) of initial shockable and 4.1% (95% CI, 3.6%, 4.7%) of initial non-shockable events (Figure 1a). Of 215 patients with initial non-shockable rhythm who received lidocaine, 111 (52%) had a documented subsequent shockable rhythm. The adjusted odds of receiving lidocaine for all events in 2018 were lower compared to 2002 (aOR 0.33; 95% CI, 0.18, 0.61; p < 0.001; Table 2). For all arrests in the adjusted model, there were no significant changes in the step (−2.1%; 95% CI, −5.9%, 1.6%; p = 0.27) or slope (0.0% per year; 95% CI, −0.9%, 1.6%; p = 0.59) after 2010 or after 2015 (Step: −1.5%; 95% CI, −8.0%, 5.0%; p = 0.65; Slope: 0.6% per year; 95% CI, −0.9%, 1.6%; p = 0.49; Figure 1b). These findings remained consistent when looking at only shockable events, only events in which at least one anti-arrhythmic was used and when 2011 was used as the first intervention year (Supplementary Table 3).
Figure 1:
A. Observed rates with 95% CI of lidocaine use over time by initial rhythm. B. Overall predicted rates with 95% CI of lidocaine use with interrupted time series overlay Vertical line indicates end of the year of guideline update (includes all of 2010 and 2015, respectively).
Table 2:
Adjusted odds ratios* with 95% CI of receiving medications in 2018 with respect to the year 2002.
| Drug | OR (95% CI) | p-value |
|---|---|---|
| Epinephrine | 0.76 (0.48–1.20) | 0.240 |
| Vasopressin | 5.19 (1.22–22.09) | 0.026 |
| Amiodarone | 0.94 (0.48–1.85) | 0.855 |
| Lidocaine | 0.33 (0.18–0.61) | <0.001 |
| Atropine | 0.19 (0.12–0.30) | <0.001 |
| Bicarbonate | 0.54 (0.35–0.86) | 0.009 |
| Calcium | 0.86 (0.62–1.21) | 0.388 |
| Magnesium | 0.54 (0.28–1.07) | 0.079 |
| Dextrose | 0.61 (0.29–1.31) | 0.211 |
Adjusted for age, monitored, witnessed, initial shockable rhythm and teaching status of hospital.
Sodium Bicarbonate
Sodium bicarbonate was used in 55.7% (95% CI, 54.4%, 56.9%) of all arrests including 59.0% (95% CI, 55.6%, 62.3%) of initial shockable and 55.1% (95% CI, 53.8%, 56.5%) of initial non-shockable events (Figure 2a). The odds of receiving bicarbonate in 2018 were lower compared to 2002 in the adjusted model (aOR 0.54; 95% CI, 0.35, 0.86; p = 0.009; Table 2). In the subgroup analysis limited to prolonged events, bicarbonate use did not change overtime (aOR 1.00; 95% CI, 0.56, 1.79; p = 0.994; Supplementary Table 4).
Figure 2:
A. Observed rates with 95% CI of sodium bicarbonate use over time by initial rhythm. B. Overall predicted rates with 95% CI of sodium bicarbonate use with interrupted time series overlay. Vertical line indicates the end of the year of guideline update (includes all of 2010).
For all arrests in the adjusted model, there were no significant changes in the step (−2.3%; 95% CI, −7.6%, 3.0%; p = 0.39) or slope (1.1% per year; 95% CI, 0.0%, 2.2%; p = 0.05) after 2010 (Figure 2b). These findings remained consistent when 2011 was used as the intervention year (Supplementary Table 3).
Atropine
Atropine was used in 27.0% (95% CI, 25.9%, 28.1%) of all arrests including 20.4% (95% CI, 17.7%, 23.3%) of initial shockable and 28.1% (95% CI, 26.9%, 29.3%) of initial non-shockable events (Figure 3a). The adjusted odds ratio for atropine use was 0.19 (95% CI, 0.12, 0.3; p < 0.001) in 2018 compared to 2002 (Table 2). For all arrests in the model, there was a downward step change after 2010 (−5.9%; 95% CI, −10.5%, −1.3%; p = 0.01), but not in the slope (−0.6% per year; 95% CI, −1.5%, 0.0%; p = 0.27; Figure 3b). This remained true when 2011 was used as the intervention year (Supplementary Table 3).
Figure 3:
A. Observed rates with 95% CI of atropine use over time by initial rhythm. B. Overall predicted rates with 95% CI of atropine use with interrupted time series overlay. Vertical line indicates the end of the year of guideline update (includes all of 2010).
The observed rates of atropine use by age group are shown in Figure 4a. In the subgroup analyses by age group, there was a substantial downward step in the use of atropine after 2010 in adolescents (−17.4%; 95% CI, −27.4%, −7.3%; p = 0.001; Figure 4b), but not for any other age group (Supplementary Table 5). However, when 2011 was used as the intervention year, substantial downward step changes were noted in neonates (−16.8%; 95% CI, (−27.0%, −6.6%; p = 0.001) and infants (−17.6%; 95% CI, (−28.4%, −6.9%; p = 0.001; Figure 4c) but not in the child age group.
Figure 4:
A. Observed rates (95% CI omitted for readability) of atropine use over time by age group. B. Overall predicted rates with 95% of atropine use with interrupted time series overlay in adolescents. Vertical line indicates the end of the year of guideline update (includes all of 2010). C. Overall predicted rates with 95% CI of atropine use with interrupted time series overlay in neonates and infants. Vertical line indicates the end of the year of publication of updated provider manual (includes all of 2011).
Other Medications
The observed rates of use of the remaining medications by initial rhythm are shown in Supplementary Figure 3. The odds of patients receiving epinephrine, amiodarone, calcium, magnesium and dextrose were similar in 2018 compared to 2002 (Table 2). Vasopressin use was higher in 2018 compared to 2002 with an adjusted odds ratio of 5.19 (95% CI, 1.22, 22.09; p = 0.026), though overall use throughout the study period remained low (9.9%; 95% CI, 7.4%, 12.9% in 2018).
DISCUSSION
In this analysis of a United States national registry of pediatric pulseless IHCA, we found that changes in PALS recommendations for lidocaine and bicarbonate were not temporally associated with acute changes in the use of these medications; however, use did decrease over the study period. In contrast, a minor addition to the guideline text regarding atropine in pediatric cardiac arrest was associated with a modest decrease in the use of atropine after 2010. Except for vasopressin, trends in use of the remaining medications over the study period remained stable despite a significant decrease in arrest duration.
Why the changes to the pediatric recommendations for intra-arrest lidocaine and bicarbonate use were not temporally associated with changes in their prescribing patterns remains unclear. However, use of both medications (and atropine) did decrease over the study, while similar trends were not seen with medications for which the guidelines did not change. This suggests that although changes did not occur acutely surrounding the guideline updates, PALS prescribers have shifted their practice over time to better align with the recommendations.
For lidocaine, the more gradual decline in use observed in our study is consistent with the patterns seen in adult cardiac arrest following a similar deemphasis in the strength of the ACLS recommendation for its use.5, 15 Importantly, both PALS and ACLS maintained presence of lidocaine as an alternative to amiodarone1, 2, 6, 7, 15–18, and therefore, continued lidocaine use during this period may have additionally been affected by other factors including cost or availability of amiodarone which we could not account for in the present study. More time may be needed to assess a more gradual change in practice after the 2015 PALS update which again includes lidocaine in the pictorial algorithm and recommends the medication as an equivalent to amiodarone. Additionally, we found that 2% of patients with initial non-shockable rhythm received lidocaine despite no documentation of subsequent shockable rhythm. Though we believe this is most likely due to documentation errors, it is possible that use of anti-arrhythmics in purely non-shockable IHCA may represent an anomaly in some providers’ practice that should be further explored.
Though its overall use declined over the study period, bicarbonate was still administered in nearly half of all arrests in 2018 despite the guideline change from being considered for prolonged arrest to recommending against its routine use in 2010.2 As we were unable to determine what proportion of events included bicarbonate for circumstances in which it has consistently been recommended (such as hyperkalemia and certain toxidromes5, 19) we cannot draw conclusions on providers’ adherence to the recommendations. However, we speculate that the overall decline in use, which was largely driven by events lasting ≤10 minutes, reflects a shift over time in providers’ strategies to refraining from routine administration. This may have been the case for shorter events, however given the additional challenges associated with prolonged IHCA, some providers may have subsequently extended their strategies beyond that of guidelines leading to the consistent use of bicarbonate observed in this subgroup.
Prior to 2010, atropine had not been mentioned in PALS for pulseless cardiac arrest, though it has consistently been recommended for vagally-meditated bradycardia with a pulse and poor perfusion. However, in the 2010 PALS guidelines, the statement “There is insufficient evidence to support or refute the routine use of atropine for pediatric [pulseless] cardiac arrest” was added. Although this did not appear to represent a major change in the recommendations for atropine in pediatric cardiac arrest, this was the only instance in our study in which a significant acute change in medication use was observed following a guideline update. Of note, the 2010 PALS update coincided with a major change in the adult ACLS guidelines in which atropine was removed from the PEA/asystole algorithm. This update was associated with a rapid and profound decrease in the use of intra-arrest atropine in adults.5
Our findings can partially be explained by the substantial drop in the use of atropine in adolescent patients after 2010. As many pediatric providers endorse following ACLS guidelines in the post-pubertal adolescent, it is likely that this subset of patients was receiving care per the ACLS guidelines. The neonatal and infant age groups also experienced a significant decrease in atropine use when 2011 was used as the intervention year. This observation is likely due to a delay from the time from publication of the PALS guidelines to the updates in the PALS Provider Manual compared to those for the ACLS educational materials. Specifically, the PALS Provider Manual was published a full year after the 2010 guidelines. 2, 20 The slower dissemination of educational materials may have led to the delay in the shift in practice patterns in infants and neonates. Though atropine use did not acutely change during the study period in children aged 1 year to <12 years, the steady decline in the use again suggests that the practice has shifted more slowly over time in accordance with available evidence. It is also notable that by 2018, 19% of all arrests still included the use of atropine. As 26% of the events in our cohort started as non-pulseless and progressed to pulseless IHCA, we speculate that this finding is at least partially explained by the appropriate administration of atropine during the non-pulseless phase of the event. Though, we are unable to confirm this theory without detailed data on timing of medication administration.
There were several limitations to our study. First, we were unable to account for other factors that may have influenced prescribing patterns such as medication cost or availability. Second, only hospitals participating in GWTG-R registry could be included in the analysis, and nearly all of these were academic hospitals. Therefore, our results may not be generalizable to all pediatric ICUs. Finally, we were unable to assess providers’ adherence to guidelines with respect to intra-arrest medications, given that the GWTG-R registry does not collect granular event data including timing of medication administration (other than epinephrine), medication dosing, timing of rhythm changes or specific causes of arrest. This additionally restricted us from correlating medication use with outcomes due to the related risk of bias.21
CONCLUSION:
Changes to the PALS guidelines for lidocaine and bicarbonate were not temporally associated with changes in the use of these medications in pediatric IHCA; however, trends in their use over time reflect better alignment with updates to the recommendations. A minor update to the language surrounding atropine in the PALS text was associated with a modest change in the use of atropine after 2010. Understanding other factors that influence prescribers in pediatric IHCA may help direct future PALS guideline development and implementation.
Supplementary Material
Supplementary Figure 1: Consort diagram.
Supplementary Figure 2: Median event duration with inter-quartile range over time.
Supplementary Figure 3: Observed rates with 95% CI of amiodarone, calcium, dextrose, epinephrine, magnesium and vasopressin use over time by initial rhythm for initial non - shockable (A) and shockable (B) events.
ACKNOWLEDGEMENTS:
Get With The Guidelines-Resuscitation Investigators: Besides the author Monica E. Kleinman, M.D., members of the Get With The Guidelines-Resuscitation Pediatric Research Task Force include: Anne-Marie Guerguerian MD PhD FRCPC; Dianne Atkins MD; Elizabeth E. Foglia MD MSCE; Ericka Fink MD; Javier J. Lasa MD FAAP; Joan Roberts MD; Jordan Duval-Arnould MPH DrPH; Melania M. Bembea MD MPH PhD; Michael Gaies MD MPH MSc; Punkaj Gupta MBBS; Robert M. Sutton MD MSCE FAAP FCCM; Taylor Sawyer DO Med
FUNDING SOURCES:
Dr. Ross received funding for this project from the American Heart Association’s Get With the Guidelines Young Investigator Database Seed Grant. Get With The Guidelines-Resuscitation provided the registry data and reviewed the research proposal and final manuscript for accuracy and scientific rigor.
Dr. Moskowitz’s work is supported by K23GM128005-01.
Dr. Donnino’s work is supported by NHLBI K24 HL127101.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
CONFLICTS OF INTEREST: Drs. Moskowitz, Grossestreuer, Kleinman and Donnino hold volunteer roles at the American Heart Association.
REFERENCES:
- [1].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:S526–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Kleinman ME, de Caen AR, Chameides L, et al. Part 10: Pediatric basic and advanced life support: 2010 International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science With Treatment Recommendations. Circulation. 2010;122:S466–515. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [3].ECC Committee and Subcommittees and Task Forces of the American Heart Association. Part 12: Pediatric Advanced Life Support. Circulation. 2005;112:IV-167–IV-87. [Google Scholar]
- [4].American Heart Association in collaboration with International Liaison Committee on Resuscitation. Guidelines 2000 for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Part 10: pediatric advanced life support. Circulation. 2000;102:I291–342. [PubMed] [Google Scholar]
- [5].Moskowitz A, Ross CE, Andersen LW, et al. Trends Over Time in Drug Administration During Adult In-Hospital Cardiac Arrest. Critical care medicine. 2018. [DOI] [PMC free article] [PubMed]
- [6].Link MS, Berkow LC, Kudenchuk PJ, et al. Part 7: Adult Advanced Cardiovascular Life Support: 2015 American Heart Association Guidelines Update for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation. 2015;132:S444–64. [DOI] [PubMed] [Google Scholar]
- [7].Neumar RW, Otto CW, Link MS, et al. Part 8: adult advanced cardiovascular life support: 2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation. 2010;122:S729–67. [DOI] [PubMed] [Google Scholar]
- [8].Honarmand K, Mepham C, Ainsworth C, Khalid Z. Adherence to advanced cardiovascular life support (ACLS) guidelines during in-hospital cardiac arrest is associated with improved outcomes. Resuscitation. 2018;129:76–81. [DOI] [PubMed] [Google Scholar]
- [9].Antonio Moretti M, Machado Cesar LA, Amit Nusbacher B Kern K, Sergio Timerman, Franchini Ramires JA. Advanced cardiac life support training improves long-term survival from in-hospital cardiac arrest. Resuscitation. 2007;72:458–65. [DOI] [PubMed] [Google Scholar]
- [10].Mcevoy MD, Field LC, Moore HE, Smalley JC, Nietert PJ, Scarbrough SH The effect of adherence to ACLS protocols on survival of event in the setting of in-hospital cardiac arrest. Resuscitation. 2014;85:82–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Peberdy MA, Ornato JP, Larkin GL, et al. Survival from in-hospital cardiac arrest during nights and weekends. Jama. 2008;299:785–92. [DOI] [PubMed] [Google Scholar]
- [12].Nadkarni VM, Larkin GL, Peberdy MA, et al. First documented rhythm and clinical outcome from in-hospital cardiac arrest among children and adults. Jama. 2006;295:50–7. [DOI] [PubMed] [Google Scholar]
- [13].American Hospital Association. AHA Annual Survey Database fiscal year 2018. https://www.ahadataviewer.com/book-cd-products/aha-survey/.
- [14].Lumley T, Diehr P, Emerson S, Chen L. The importance of the normality assumption in large public health data sets. Annu Rev Public Health. 2002;23:151–69. [DOI] [PubMed] [Google Scholar]
- [15].The American Heart Association in collaboration with the International Liaison Committee on Resuscitation. Guidelines 2000 for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Part 6: advanced cardiovascular life support: 7D: the tachycardia algorithms. Circulation. 2000;102:I158–65. [PubMed] [Google Scholar]
- [16].Ecc Committee S, Task Forces of the American Heart A. 2005 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation. 2005;112:IV1–203. [DOI] [PubMed] [Google Scholar]
- [17].ECC Committee, Subcommittees and Task Forces of the American Heart Association. 2005 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care: Part 12: Pediatric Advanced Life Support. Circulation. 2005;112:IV1–IV203. [DOI] [PubMed] [Google Scholar]
- [18].The American Heart Association in collaboration with the International Liaison Committee on Resuscitation. Guidelines 2000 for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Part 10: pediatric advanced life support. Circulation. 2000;102:I291–342. [PubMed] [Google Scholar]
- [19].American Academy of Pediatrics, American Heart Association. Pediatric Advanced Life Support Provider Manual. Dallas, TX: American Heart Association; 2016. [Google Scholar]
- [20].Chameides L, Samson RA, Schexnayder SM, Hazinski MF. Pediatric Advanced Life Support Provider Manual. First ed. Dallas, TX: American Heart Association; 2011. [Google Scholar]
- [21].Andersen LW, Raymond TT, Berg RA, et al. Association Between Tracheal Intubation During Pediatric In-Hospital Cardiac Arrest and Survival. Jama. 2016;316:1786–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Figure 1: Consort diagram.
Supplementary Figure 2: Median event duration with inter-quartile range over time.
Supplementary Figure 3: Observed rates with 95% CI of amiodarone, calcium, dextrose, epinephrine, magnesium and vasopressin use over time by initial rhythm for initial non - shockable (A) and shockable (B) events.