a b s t r a c t

Objectives: To investigate the effect of medical student involvement on the quality of actual cardiopulmonary resuscitation (CPR).

Methods: A digital video-recording system was used to record and analyze CPR procedures for adult patients from March 2011 to September 2012.

Results: Twenty-six student-involved and 40 non-student-involved cases were studied. The chest compression rate in the student-involved group was significantly higher than that in the non-student- involved group (P b .001). The proportion of compressions at “above 110 cpm” was higher in the student- involved group (P = .021), whereas the proportion at “90-110 cpm” was lower in the student-involved group (P = .015). The ratio of hands-off time to total Manual compression time was significantly lower in the student-involved group than in the non-student-involved group (P = .04). In contrast, the student-involved group delivered a higher ventilation rate compared with the non-student-involved group (P = .02). The observed time delay to first compression and first ventilation were very similar between the groups. There were no significant differences between the groups in either return of spontaneous circulation or time from survival to discharge.

Conclusion: Student-involved resuscitation teams were able to perform good CPR, with higher compression rates and fewer interruptions. However, the supervision from medical staff is still needed to ensure appropriate chest compression and ventilation rate in student-involved actual CPR in the emergency department.

(C) 2014

1. Introduction

Clinical rotations are designed to introduce medical students to the basic cognitive and technical skills for the practice of medicine. However, in some circumstances, students do not receive adequate exposure to clinical problems and lack the opportunity to acquire medical skills during rotations [1,2]. This may partially due to the fact that supervising physicians cannot put a patient at risk when a student appears to be inadequate in providing medical services, and many patients do not want a medical student to perform a procedure on them regardless of the student’s level of training [3,4]. This creates a dilemma, especially for teaching hospitals. Although it seems intuitively reasonable that a student-involved procedure may pro- voke more unexpected errors and may alter the medical care quality, the real situation is unknown. Profiting from a special circumstance,

? Conflicts of interest statement: The authors do not have any conflict of interests to disclose.

* Corresponding author. Emergency Center, Zhongnan Hospital of Wuhan University, Wuchang, Wuhan, Hubei 430071, China. Tel.: +86 027 67813337.

E-mail address: [email protected] (Y. Zhao).

we determined whether the Quality of cardiopulmonary resuscitation (CPR) was altered by the involvement of medical students in our emergency department (ED).

1. Methods
   1. Patient enrollment

This study was conducted from March 2011 to September 2012, in a 30-bed ED, with an annual census of 50 000 in a teaching hospital affiliated to Wuhan University. Both out-of hospital and in-hospital patients older than 18 years with cardiac arrest from any cause admitted to our resuscitation room were eligible for inclusion. Only those CAs occurred in the ED were included as “in-hospital patients” in this study. The CA patient inclusion was done via a search of our electronic medical records system. True arrest cases were defined by apnea, documented loss of pulse and consciousness, and the delivery of chest compression.

All patients with CA admitted to our rescue room will receive CPR. Our ED has well-trained resuscitation teams including basic life support (BLS) and Advanced cardiovascular life support –

0735-6757/$ – see front matter (C) 2014 http://dx.doi.org/10.1016/j.ajem.2013.12.007

capable and certified staff. Our staff has rich experiences in performing CPR for patients with CA. The ACLS physicians also occasionally assist in performing CPR for patients with CA in other inpatient departments. As an American Heart Association training center, our department provides opportunities for us to organize regular theoretical knowledge update and skill trainings for our students and medical staff. In our ED, CPR was directed by physicians certified in ACLS with assistance from nurses, medical students, and respiratory technicians. All potential rescuers were at least BLS certified. All fourth-year medical students in our school should participate in a 4-week ED rotation, which has been done in a longitudinal manner throughout the year. Those students are primarily assigned to the rescue room of the ED. About 2 to 5 medical students are in charge of every day and night shift in our department. Students work closely with our residents while providing direct patient care. All patient care is directly and closely supervised.

The ethics committee of Zhongnan Hospital of Wuhan University (Wuhan, Hubei, China) approved this study. Waived informed consent was authorized because the study did not interfere with Standard care provided to the patients. Nevertheless, information about video recording was provided to the relatives of patients. All medical staff and students sign the informed consent forms for the study. Our study does not interfere with the clinical exposures or opportunities for medical students. Besides, we did not interfere with any Clinical decisions of our medical staff.

Video recording and CPR quality measurement

In the resuscitation room of our department, a video-recording system (DS-8000 video network hard disk recorder; Hikvision Company, Hangzhou, China) was installed, which recorded all events occurring in the resuscitation room (Fig. 1). All events were automatically time stamped and stored on a hard disk. We transferred the CPR videos into a mobile hard disk for analysis.

Clinical data including patients’ demographics and event charac- teristics were obtained from their ED medical record, in-patient medical record, autopsy data, written resuscitation transcripts, and real-time video records. For each patient, the following data were collected: age, sex, initial cardiac rhythm, causes of CA, medical

Fig. 1. A typical CPR scene obtained from camera number 5 in our rescue room.

history, time of resuscitation in a day, number of members of the CPR team, number of students involved in resuscitations, use of automatic chest compression device, return of spontaneous circulation (ROSC), and survival to discharge.

We only analyzed the first 30 minutes when total CPR time exceeded 30 minutes. If the resuscitation team used an automatic chest compression device ZOLL AutoPulse (Model 100; Zoll Circula- tion, Chelmsford, MA) or Thumper CPR (Model 1007; Michigan Instruments, Inc, Grand Rapids, MI) after a period of Manual CPR, we only analyzed the manual CPR period. The main measures of CPR quality were considered as follows: (1) instantaneous rates of chest compression, (2) hands-off time, (3) intubation time, and (4) rates of ventilation with Bag-valve-mask . According to our previous study [5], the recordings were divided into 30-second segments for calculating instantaneous rates from the following formula: rate = (compressions per 30-second segment) x 60/(30 – total pause time in 30-second segment(s)), where pause time was defined as periods not shorter than 1 second without chest compressions, including time to check pulses or to deliverer a shock. We recorded the causes and the length of hands-off and calculated hands-off ratio of hands-off time to total manual compression time. Time delay was also recorded. Time delay in in-hospital arrest was defined as the time from witnessed arrest or the time from vital sign monitoring system-recorded arrest to first chest compression or ventilation, whereas time delay in out-of hospital arrest was defined as the time lag from arrival of the patient at the resuscitation room to the first compression or ventilation.

All enrolled CPR videos were reviewed by 2 independent

observers, a physician with BLS and ACLS certification and a graduate student with CPR knowledge and skills. These 2 observers have not participated in any enrolled case of this study. Both observers were trained to watch videos in a same standard mode and entered information onto a previously designed data sheet. A third expert physician was consulted to give the final decision when a disagree- ment occurred. Cases were divided into student-involved group and non-student-involved group. In the student-involved group, students should have actively participated in CPR. Otherwise, the arrest cases have been defined as non-student-involved ones.

Statistical analysis

Data were expressed as mean +- SD, median (25%-75% inter- quartile range), and number or percentages. Evaluations of CPR variable distributions were performed using a Bartlett homogeneity of variances. Comparison of means was performed using the 2-tailed t test or separate variance estimation t test; comparison of medians was performed using the Wilcoxon-Mann-Whitney test; and comparison of percentages was performed using the ?² test or the Fisher exact method, if appropriate. Statistical analysis was performed using SPSS software (SPSS 18.0; SPSS Inc, Chicago, IL). A 2-tailed P value of less than .05 was considered significant.

1. Results

In March 2011 to September 2012, a total of 70 CPR cases met the inclusion criteria of the study. Because of unclear scenes on video screen or device dysfunction, we lost 4 cases. Therefore, 66 consecutive cases were divided into student-involved group (26 cases) and non-student-involved group (40 cases) for analysis. The demographic characteristics of patients are shown in the Table 1. Differences between groups were not statistically significant. Arrests also occurred in similar time-of-day distributions in 2 groups. The number of team members in the student-involved group was significantly larger than that in the non-student-involved group (7 [5-9] vs 5 [4-7], P = .04). Finally, both the student-involved and non- student-involved groups had similar distribution of presenting CA

Table 1

Patient demographics and clinical resuscitation data

	Student group (n = 26)	Non-student group (n = 40)	P
Age (y)	57 +- 10	56 +- 9	.54
Sex, male	18 (69)	30 (75)	.39
Patient classification Internal medicine	20 (77)	32 (80)	.81
Trauma Time of events Morning (6 AM-12 PM)	6 (23) 7 (27)	8 (20) 11 (27)	.98
Afternoon (12-6 PM)	8 (31)	13 (34)
Evening (6 PM-12 AM)	6 (23)	8 (20)
Night (12-6 AM) Initial rhythm PEA	5 (19) 10 (39)	8 (20) 14 (35)	.32
VF/VT	0 (0)	2 (5)
Asystole	12 (46)	16 (40)
Unknown History Known	4 (15) 18 (69)	8 (20) 30 (75)	.88
Unknown Cause of CA Cardiogenic and pulmonary causes	8 (31) 10 (38)	10 (25) 22 (55)	.40
Hemorrhage causes	4 (15)	6 (15)
Intoxications	2 (8)	4 (10)
Unknown	10 (38)	8 (20)
No. of team member	7 (5-9)	5 (4-7)	.04
Doctors/Nurses	4 (4-6)	5 (4-7)	.19
Students Use of chest compression device	2 (1-3)	0	NA
AutoPulse (Zoll)	10 (38)	19 (48)	.44
Thumper 1008	8 (31)	12 (30)

Data are mean +- SD, median (25%-75% interquartile), or number (percentage). “Known” means exact medical record or knowledge about preexisting diseases according to family or companion.

Abbreviations: NA, unavailable or not carried out; PEA, pulseless electrical activity; VF, ventricular fibrillation; VT, ventricular tachycardia.

rhythms, with most patients in both groups presenting with asystole and pulseless electrical activity.

A total of 54 students were involved in this study. The characteristics of students are summarized in Table 2. Most students participated in chest compression and ventilation with BVM; the others was engaged in defibrillation attempts, assessment of patient response, and observation of vital signs. In addition, of 92% (24/26 cases) student-involved cases, both ventilation with BVM and manual chest compression were partially delivered by students.

Instantaneous rates of chest compression showed a marked variation in the 66 cases (Fig. 2). During the total of 2320 analyzed 30-second periods in 2 groups, the proportions of chest compression rates at “below 90 cpm,” “90-110 cpm,” and “above 110 cpm” were 15%, 23%, and 62% in the student-involved group and 20%, 32%, and 48% in the non-student-involved group, respectively (Fig. 2). The proportion of inappropriate compression rates (below 90 cpm) was not different (15% vs 20%, P = .13). However, the proportion of 90 to

Table 2

Characteristics of students (n = 54)

Age (y), mean (range) 22 (21-23)

Sex, male	30/54 (56)
Grade (year)
4th	38/54 (70)
5th	16/54 (30)
BLS certified (yes)	54/54 (100)
Procedure delivered
Chest compression	52/54 (96)
Ventilation with BVM	44/54 (82)
Defibrillation	2/54 (4)
Pulse/Rhythm analysis	8/54 (15)
Others	6/54 (11)

Data are number (percentage), unless otherwise indicated.

110 cpm and above 110 cpm was significantly different between groups (32% vs 23% [P = .015] and 48% vs 62% [P = .021], respectively). There was no significant difference between groups in time delay to first chest compression. Chest compression rate fell from 121 +- 33 min^-1 in the student-involved group to 107 +- 35 min^-1 in the non-student-involved group (Table 3, P b .001).

The ratio of hands-off time to total manual compression time was different in 2 groups. In the student-involved group, hands-off time ratio was (11% +- 3%) during resuscitation, which is significantly lower than that in the non-student-involved group (24% +- 6%; Table 3, P =

.04), suggesting less frequent pauses in CPR or less frequent rescuer switching occurred in the student-involved group, which may relate to physical factors or number of team numbers. The provider switches were the most common causes for chest compression interruption in both the student-involved and non-student-involved groups. The ventilation rate before intubation was significantly higher in the student-involved group (22 +- 10 breaths/min vs 12 +- 5 breaths/min, P = .02). The intubation time was very similar for both groups (140 +- 33 seconds in the student-involved group vs 133 +- 28 seconds in the non-student-involved group, P = .21). There was no significant difference between groups in time delay to first ventilation and time lag from patient arrived to first in-hospital intravenous pathway completed (Table 3).

X.-L. Zhou et al.Patient outcomes are shown in Table 3. A small, statistically insignificant increase in the rate of ROSC was seen between student-involved and non-student-involved group (35% vs 31%, P = .70). There was no significant difference in survival to discharge between groups.

1. Discussion

To the best of our knowledge, this is the first study to specifically evaluate the involvement of medical student on the quality of actual CPR occurred in ED. The higher compression rate and less hands-off time compared with non-student-involved teams indicated that student-involved resuscitation teams in our ED performed good- quality CPR.

The Quality of CPR is an important determinant of outcome after CA [6-9]. A high-quality CPR is considered as chest compression of adequate depth and rate with complete chest wall release in between compressions, minimal hands-off time, and appropriate ventilation [10]. Numerous studies indicated that providing high-quality chest compression with minimal interruptions might be one of the most important actions during CA that will translate into a Survival benefit [11-13]. During CPR resuscitation, chest compression rate should be delivered at a rate of at least 100 compressions/min with a compression depth of at least 5 cm, according to American Heart Association guidelines 2010. However, many studies suggested that the conventional manual chest compression was often performed ineffectively [14,15]. Studies in both the in-hospital CA and out-of- hospital CA settings have shown that compression rates are lower than the international guidelines-recommended rate [14-17]. In both human and animal studies, minimal interruptions of chest compres- sion appear to be a very important factor in determining CPR quality and outcome [18-21]. Unfortunately, interruptions in chest compres- sion are common. Wik et al [15] found that patients with out-of- hospital CA received compression only about half of the time, and Stiell et al [22] found suboptimal compression depth in half of patients by 2005 guideline standards and almost all by 2010 standards. In our study, student-involved resuscitation groups performed CPR with a higher chest compression rate and less hands-off time ratio. This may be explained by the differences in physical strength and number of team members between groups. First, although our physicians were experienced and well trained in CPR, we might speculate that the physical capacity of young medical students was better than our physicians. Second, chest compression is a high workload procedure

Fig. 2. Frequency distribution of instantaneous chest compression rates. Each value corresponds to a 30-second period of CPR (n = 920 periods obtained in 26 patients in the student-involved group and n = 1400 periods obtained in 40 patients in the non-student-involved group).

and patient visits in ED are unpredictable. It often occurs that several critically ill patients arrive at the same time, and we do not have a sufficient number of staff to take care of them. In this study, compared with the non-student-involved group, the number of team members in student-involved group was significantly higher owing to the involvement of medical students, which could ensure a sufficient number of team members to provide CPR. However, compression rate in the student-involved group might be too high and unsteady. Importantly, a high compression rate might influence chest recoil and blood flow back to the heart, which could reduce Coronary and cerebral perfusion. In addition, a recent manikin study suggested that the mean compression depth was below the recommended guideline depth for all compression rates above 80 min^-1 [23].

Positive pressure ventilation during CPR increases intrathoracic pressure, reducing venous return and cardiac output. O’Neill and Deakin [24] found that hyperventilation was common and mostly through high respiratory rates. Animal studies also demonstrated that excessive ventilation rates resulted in significantly increased intra- thoracic pressure and decreased coronary perfusion pressures and survival rates [25]. In addition, reducing the ventilation rate during CPR has been shown to improve vital organ perfusion in animal

studies [26]. In our study, student-involved resuscitation teams performed a high ventilation rate before intubation completed, which may lead to hyperventilation and subsequent reduced cardiac output. Emergency department is a very stressful environment for the medical students, and it is difficult for them to maintain an appropriate rate to ventilate with BVM.

We recognize several limitations of this study. First, only a small number of cases in one ED were enrolled in this study, and the limitations of small sample size and a lack of generalizability must be considered. In addition, this study was not powered to detect clinical outcome differences such as ROSC or survival to hospital discharge. Outcomes evaluation will require larger trials to be performed. Second, the data in our study were collected via 2 different observers, so human error might affect our findings. We have attempted to address this concern in several ways. The observers were trained to watch videos in a same standard mode, and a third expert physician was consulted to give the final decision when an assessment of one recording was different between 2 observers. Third, although all potential rescuers were at least BLS certified and physicians were certified in ACLS, we are limited in our ability (via a video-recording system or medical records) to analyze other important variables that

Table 3

CPR quality comparison

	Student group (n = 26)	Non-student group (n = 40)	P value (means)	P value (variance)
Instantaneous CC rates (cpm)	121 +- 33	107 +- 35	b.001	.47
Hands-off time ratio (%)	11 +- 3	24 +- 6	b.01	.04
Total CC interruptions	278	612	NA	NA
Causes of total CC interruptions
Provider switch	139/278 (50%)	245/612 (40%)	.92
Pulse/Rhythm analysis	56/278 (20%)	122/612 (20%)
Airway management	28/278 (10%)	123/612 (20%)		NA
Defibrillation attempts	27/278 (10%)	61/612 (10%)
Others/Undetermined	28/278 (10%)	61/612 (10%)
Time delay to first CC (s)	10 +- 4	12 +- 6	.49	.048
Time delay to first ventilation (s)	69 +- 14	77 +- 20	.32	.021
Ventilation rate with BVM (min-1)	22 +- 10	12 +- 5	.02	.055
Intubation time (s)	140 +- 33	133 +- 28	.21	.071
Time lag to first IV completed (s)	410 +- 74	388 +- 55	.42	.003
ROSC	8/26 (31%)	14/40 (35%)	.70	NA
Survival to discharge	2/26 (8%)	4/40 (10%)	.18	NA

Data are mean +- SD or number (percentage).

Abbreviations: CC, chest compression; NA, unavailable or not carried out.

may effect CPR quality such as real training level (students), years of experience in profession (doctors/nurses), provider cardiovascular fitness, or physical strength. Finally, and perhaps the most important limitation, we did not measure chest compression depth and incomplete decompression in this study. Stiell et al [22] found a strong association between survival outcomes and increased com- pression depth. In addition, incomplete decompression had a detrimental effect on coronary perfusion pressure and cerebral perfusion pressure, which, in turn, is a primary predictor of survival [27]. Abella et al [14] reported that one-third of compressions were too shallow for CPR after in-hospital CA. Meanwhile, the study of Wik et al [15] showed that 59% of compressions were too shallow and no- flow times approached 50%.

1. Conclusion

This observational study suggested that student-involved resusci- tation teams performed good CPR, with higher compression rates and fewer interruptions. However, the chest compression rate might be too high to ensure appropriate compression depth and complete chest wall release, and high rate of ventilation before intubation completed in student-involved teams might bring adverse effects due to possible hyperventilation. Thus, supervision from physicians is necessary for medical students who are involved in actual CPR in the ED.

Acknowledgments

The authors would like to thank the medical staffs in the ED of Zhongnan Hospital and the medical students from Wuhan University. Their efforts made this study possible.

References

1. Fincher RM, Lewis LA. Learning, experience, and self-assessment of competence of third-year medical students in performing bedside procedures. Acad Med 1994;69:291-5.
2. Taylor DM. Undergraduate procedural skills training in Victoria: is it adequate? Med J Aust 1997;166:251-4.
3. Graber MA, Wyatt C, Kasparek L, et al. Does simulator training for medical students change patient opinions and attitudes toward medical student pro- cedures in the emergency department? Acad Emerg Med 2005;12:635-9.
4. Graber MA, Pierre J, Charlton M. Patient opinions and attitudes toward medical student procedures in the emergency department. Acad Emerg Med 2003;10: 1329-33.
5. Jiang C, Zhao Y, Chen Z, et al. Improving cardiopulmonary resuscitation in the emergency department by real-time video recording and regular feedback learning. Resuscitation 2010;81:1664-9.
6. Wik L, Steen PA, Bircher NG. Quality of bystander cardiopulmonary resuscitation influences outcome after prehospital cardiac arrest. Resuscitation 1994;28: 195-203.
7. Berg RA, Hilwig RW, Kern KB, et al. “Bystander” chest compressions and assisted ventilation independently improve outcome from piglet asphyxial pulseless “cardiac arrest”. Circulation 2000;101:1743-8.
8. Ko PC, Chen WJ, Lin CH, et al. Evaluating the quality of prehospital cardiopulmo- nary resuscitation by reviewing automated external defibrillator records and survival for out-of-hospital witnessed arrests. Resuscitation 2005;64:163-9.
9. Edelson DP, Litzinger B, Arora V, et al. Improving in-hospital cardiac arrest process and outcomes with performance debriefing. Arch Intern Med 2008;168:1063-9.
10. Berg RA, Hemphill R, Abella BS, et al. Part 5: adult basic life support: 2010 American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation 2010;122:S685-705.
11. Olasveengen TM, Sunde K, Brunborg C, et al. Intravenous drug administration during out-of-hospital cardiac arrest: a randomized trial. JAMA 2009;302:2222-9.
12. van Walraven C, Stiell IG, Wells GA, et al. Do advanced cardiac life support drugs increase resuscitation rates from in-hospital cardiac arrest? The OTAC Study Group. Ann Emerg Med 1998;32:544-53.
13. Sutton RM, Nadkarni V, Abella BS. “Putting it all together” to improve Resuscitation quality. Emerg Med Clin North Am 2012;30:105-22.
14. Abella BS, Alvarado JP, Myklebust H, et al. Quality of cardiopulmonary resuscitation during in-hospital cardiac arrest. JAMA 2005;293:305-10.
15. Wik L, Kramer-Johansen J, Myklebust H, et al. Quality of cardiopulmonary resuscitation during out-of-hospital cardiac arrest. JAMA 2005;293:299-304.
16. Committee ECC. Subcommittees and task forces of the American Heart Association. 2005 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 2005;112: IV1-IV203.
17. Abella BS, Sandbo N, Vassilatos P, et al. Chest compression rates during cardiopulmonary resuscitation are suboptimal: a prospective study during in- hospital cardiac arrest. Circulation 2005;111:428-34.
18. Edelson DP, Abella BS, Kramer-Johansen J, et al. Effects of compression depth and pre-shock pauses predict defibrillation failure during cardiac arrest. Resuscitation 2006;71:137-45.
19. Yu T, Weil MH, Tang W, et al. Adverse outcomes of interrupted precordial compression during automated defibrillation. Circulation 2002;106:368-72.
20. Berg RA, Sanders AB, Kern KB, et al. Adverse hemodynamic effects of interrupting chest compressions for rescue breathing during cardiopulmonary resuscitation for ventricular fibrillation cardiac arrest. Circulation 2001;104:2465-70.
21. Valenzuela TD, Kern KB, Clark LL, et al. Interruptions of chest compressions during emergency medical systems resuscitation. Circulation 2005;112:1259-65.
22. Stiell IG, Brown SP, Christenson J, et al. What is the role of chest compression depth during out-of-hospital cardiac arrest resuscitation? Crit Care Med 2012;40: 1192-8.
23. Field RA, Soar J, Davies RP, et al. The impact of chest compression rates on Quality of chest compressions–a manikin study. Resuscitation 2012;83:360-4.
24. O’Neill JF, Deakin CD. Do we hyperventilate cardiac arrest patients? Resuscitation 2007;73:82-5.
25. Aufderheide TP, Sigurdsson G, Pirrallo RG, et al. Hyperventilation-induced hypotension during cardiopulmonary resuscitation. Circulation 2004;109: 1960-5.
26. Yannopoulos D, Tang W, Roussos C, et al. Reducing ventilation frequency during cardiopulmonary resuscitation in a porcine model of cardiac arrest. Respir Care 2005;50:628-35.
27. Yannopoulos D, McKnite S, Aufderheide TP, et al. Effects of incomplete chest wall decompression during cardiopulmonary resuscitation on coronary and cerebral perfusion pressures in a porcine model of cardiac arrest. Resuscitation 2005;64: 363-72.