Several studies have compared CPR quality to evaluate the effectiveness of video-based emergency dispatch systems. However, only two retrospective observational studies reported in Korea [6,7] compared patient survival as an outcome. The implementation of video-based emergency dispatch systems compared to that of the conventional audio-based systems improved the survival-to-discharge rate (22.3% vs. 10.7%; odds ratio [OR], 2.33; P<0.001) and the rate of good neurological outcomes at discharge (16.0% vs. 6.3%; OR, 2.77; P<0.001) [8]. The 2020 Korean CPR Guidelines recommend dispatcher-assisted CPR; however, only the use of speaker phones or hands-free functions is recommended during dispatcher-assisted CPR. We recommend implementing a video-based dispatch system for improving the survival rate and neurological outcomes of patients with cardiac arrest.

In an observational study reported from North America [9], the intra-arrest transport group compared with the on-site CPR group had a lower survival-to-discharge rate (4.0% vs. 8.5%; risk difference, 4.6%; range, 4.0%–5.1%) and poor neurological outcomes (2.9% vs. 7.1%; risk difference, 4.2%; range, 3.5%–4.9%). However, it is difficult to generalize this result because the prehospital emergency medical services systems in North America differ from those in Korea in terms of the composition of emergency medical personnel, level of prehospital emergency care, and related legal regulations. The ILCOR suggested that CPR should be performed on-site, except in cases where the need for transport is clear, such as in cases of extracorporeal membrane oxygenation (ECMO) candidates.

The 2020 Korean CPR Guidelines recommend that transfer to the hospital be considered if spontaneous circulation is not restored after 6 minutes of basic life support or 10 minutes of advanced life support at the scene. For CPR during transport, the Guideline Committee decided to maintain current recommendations. During transport to a hospital, high-quality CPR should be maintained.

In a retrospective observational study [10] comparing a group that received rescue ventilation for 1 minute in water when drowning and another that did not, initial survival (94.7% vs. 37.0%, P<0.001), survival-to-discharge (87.5% vs. 25%, P=0.005), and good neurological outcome (52.6% vs. 7.4%, P=0.001) rates were higher in the group that received rescue ventilation. In-water resuscitation can be considered in situations where adequately trained personnel can use appropriate equipment.

In the 2020 Korean CPR Guidelines, there was no recommendation for the use of ECMO in individuals who experienced drowning. The ILCOR analyzed two retrospective observational studies [11,12] and 11 case reports and stated that ECMO can be considered in some patients with cardiac arrest who do not respond to conventional CPR. Therefore, ECMO may be beneficial in patients with cardiac arrest or severe respiratory failure due to drowning.

Nine studies were reviewed; however, none compared the order of CPR in cases of drowning [10,13,14]. The ILCOR suggested, with expert consensus, starting CPR with chest compressions (the C [circulation]-A [airway]-B [breathing] approach) for the layperson, whereas healthcare providers could consider starting rescue ventilation before chest compressions (the A-B-C approach). The 2020 Korean Guidelines do not include recommendations on whether the CPR sequence should be changed depending on circumstances, including drowning. In addition to the suggestion of ILCOR, considering the simplicity and practicality of CPR training, the C-A-B sequence, starting with chest compressions, is recommended for the sequence of CPR even for a drowning victim.

In a targeted temperature management (TTM) trial [15] comparing target temperatures of 33 °C versus 36 °C and the TTM2 trial [16] comparing 33 °C versus normothermia below 37.5 °C (fever prevention), no difference was observed in survival to discharge, 180-day survival, or neurological outcomes at 180 days. According to a network meta-analysis [17] comparing the effects of target temperatures in 10 randomized controlled studies, compared to normothermia (37–37.8 °C), body temperatures of 31–32, 33–34, and 35–36 °C did not improve survival or neurological outcomes. The ILCOR recommended maintaining the body temperature below 37.5 °C because it can reduce the burden on medical personnel and side effects and the use of fever prevention instead of maintaining normothermia [18]. In the TTM2 study, more than 90% of the cardiac arrests were witnessed, and 62% to 63% of initial rhythms were ventricular fibrillation. The characteristics of out-of-hospital cardiac arrest (OHCA) patients in Korea are different from those of patients included in the TTM and TTM2 studies. Compared to patients enrolled in the TTM and TTM2 studies, Korean OHCA patients had a much lower incidence of a shockable rhythm and a longer time from collapse to recovery of spontaneous circulation, which might be associated with a higher chance of a severe post-cardiac arrest syndrome. Therefore, it is necessary to consider differences in the characteristics of cardiac arrest patients in Korea. In the fever prevention group of the TTM2 study [16], acetaminophen and a method of removing clothes and lowering the room temperature were used; temperature control devices were used to control body temperature in 46% of cases. Retrospective studies have suggested that hypothermia can improve neurological outcomes in patients with severe cerebral ischemic injury [19–21]. We recommend that target temperature should be maintained at 32–36 °C for at least 24 hours, as recommended by the 2020 Korean CPR Guidelines.

Three randomized controlled trials [22–24] including patients with in-hospital cardiac arrest showed that the use of vasopressin and steroids did not improve the survival to discharge rate (OR, 1.39; 95% confidence interval [CI], 0.90–2.14) or rate of good neurological outcomes (OR, 1.64; 95% CI, 0.99–2.72). However, it increased the rate of return of spontaneous circulation (ROSC; OR, 2.09; 95% CI, 1.54–2.84). The ILCOR recommended not using the vasopressin and steroid combination therapy during CPR because it has not been associated with any significant difference in survival, and no study including patients with OHCA has been conducted. The 2020 Korean Guidelines did not provide any recommendations on the combined use of vasopressin and steroids during CPR. Considering that the combination of these drugs increases the ROSC rate and that these drugs are commonly used in hospitals, we suggest vasopressin and steroid combination therapy during CPR in an in-hospital cardiac arrest setting.

There have been three observational studies; one cross-sectional study; and several case reports on pain, anxiety, agitation, and posttraumatic stress disorder in conscious patients during CPR. The studies reported that they verbally reassured the patients or administered sedatives or neuromuscular blockers [25–49]. Following the results of the scoping review conducted by the ILCOR, we recommend the following: (1) sedatives, analgesics, or both may be used in very small doses, if possible, to reduce pain and suffering in conscious patients during CPR; (2) neuromuscular blocking drugs should not be administered alone to conscious patients; (3) the optimal drug regimen for sedation and analgesia during CPR is unclear, and a regimen commonly used in critically ill patients can be used.

Twenty adult and 12 pediatric cases of cardiac arrest in the prone position have been reported. Most of these cases were observed in the operating room and one case in the intensive care unit [50–68]. There was no significant difference in the rate of ROSC or survival discharge among patients for whom chest compressions were started immediately in the prone position compared with patients for which chest compressions were started after changing the position from prone to supine. Arterial pressure during CPR was higher in the prone position group [69,70]. The end-tidal carbon dioxide partial pressure in five adults and two children was 10 mmHg or higher, and the time interval from collapse to defibrillation was shortened when defibrillation was performed in the prone position [54–56,61,65,71–73]. As recommended by ILCOR, we recommend the following for adults with cardiac arrest in the prone position: (1) if cardiac arrest occurs in the prone position with a secured airway, CPR can be started if change to the supine position is not possible or poses a significant risk to the patient; (2) if cardiac arrest occurs in the prone position without a secure airway, the patient position should be changed to supine and CPR should be initiated as soon as possible; (3) if a patient with cardiac arrest is in the prone position and cannot immediately be placed in the supine position, defibrillation can be attempted in the prone position.

One randomized controlled trial [74] and 11 cohort studies [75–85] of pediatric early warning scoring systems or implementation of a pediatric rapid response system were reviewed. The use of the pediatric early warning score tended to reduce the incidence of in-hospital cardiac arrest, mortality, and unexpected clinical deterioration. However, since research on the elements to be included in the pediatric early warning scoring system is lacking, a scoring system that can detect early warning signs of pediatric cardiac arrest, considering the available resources of each hospital, is recommended. Operation of a pediatric rapid response team is associated with a considerable decline in the preintervention trajectory of critical deterioration and a decreased likelihood of respiratory and cardiopulmonary arrest outside of the critical care unit. However, considering that the medical resources and hospital environment of the hospital where the study was conducted were different from those in Korea, the experts decided to make recommendations on the pediatric rapid response team after additional research results in Korea are released.

According to a systematic review [86] comparing delayed cord clamping and early cord clamping in preterm infants with a gestational age less than 34 weeks, delayed cord clamping resulted in significantly higher hemoglobin and hematocrit within 24 hours after birth and hematocrit at 7 days after birth and lowest mean arterial pressure within 12 hours after birth. Furthermore, the risk of using inotropics due to hypotension and blood transfusions within 24 hours after birth was significantly lower in these infants. When intact cord milking and early cord clamping were compared, hemoglobin and hematocrit levels within 24 hours after birth were significantly higher and the risk of using inotropics due to hypotension and blood transfusions within 24 hours after birth was significantly lower in infants with intact cord milking. On the other hand, comparison of delayed cord clamping and intact cord milking revealed no significant intergroup differences in these parameters. Based on these findings, the 2021 ILCOR CoSTR suggests delayed cord clamping for more than 30 seconds in preterm infants less than 34 weeks of gestational age who do not require immediate resuscitation and intact cord milking as a reasonable alternative to delayed cord clamping in preterm infants between 28 and 33 weeks of gestation. The 2020 Korean CPR Guidelines recommend that umbilical cord milking not be performed in infants less than 28 weeks of gestation due to the increased risk of intraventricular hemorrhage.

We suggest delayed cord clamping of 30 seconds or longer in preterm infants less than 34 weeks of gestation who do not require resuscitation and intact cord milking as a reasonable alternative to delayed cord clamping in preterm infants between 28 and 33 weeks of gestation.

Pre-arrest clinical prediction rules, such as pre-arrest morbidity, prognosis after resuscitation, and good outcome following attempted resuscitation scores, have been studied for predicting the prognosis of patients with in-hospital cardiac arrest; however, reliable results have not been achieved [87–93]. Therefore, the use of these predictive scales in patients with in-hospital cardiac arrest is not recommended. In addition, since there have been no related studies in pediatric patients, no recommendations can be made regarding the use of prognostic predictive scales for children with in-hospital cardiac arrest.

Potential rescuers, such as family members of high-risk patients, are less likely to voluntarily participate in CPR training but are willing to receive training [94,95]. Several studies [94–99] have recommended that basic life support be taught to potential rescuers of high-risk patients for OHCA, and that emergency staff should encourage potential rescuers to participate in basic life support. We recommend that basic life support be taught to potential rescuers of the population at high risk of cardiac arrest.

Blended learning is an educational method that combines face-to-face and non–face-to-face forms and was introduced in the 2020 Korean CPR Guidelines. Following the COVID-19 pandemic and the development of information technology, the use of various non–face-to-face methods has become more common in medical education [100–102]. Therefore, if resources and conditions allow, it is recommended to develop and implement a blended form of CPR education in addition to the conventional training methods.

Continuing CPR education for the public and emergency medical providers is important to increase survival rates after cardiac arrest. To provide continuous CPR training, an instructor training curriculum to teach trainees is important. Although many studies have explained the necessity of CPR instructor training programs, no study has reported that patient outcomes improved with the introduction of the instructor training program [103,104]. Nevertheless, because the instructor training program is an important factor in the teaching method and performance of trainees, it should be introduced.