Resuscitation, 20 (1990) 31-39 31 Elsevier Scientific Publishers Ireland Ltd.

Cardiopulmonary resuscitation skills of medical

Mohamed A. Seraj** and Mohamed Naguib

This study examined the theoretical knowledge and practical skills of different medical profession personnel and medical students in Basic Cardiac Life Support (BCLS). Two hundred twenty-four candidates who attended a BCLS course were tested in their theoretical knowledge of BCLS with 15 multiple choice questions (MCQS) as a pre-test. The same questions were included in the post-test but with a different sequence. We also evaluated each candidate during performance of single-rescuer CPR on a recording manikin, using a checklist and the recording strip from the manikin for evaluation of CPR steps and manikin performance respectively. Anesthesiologists and cardiologists demonstrated the best performance in the theoretical knowledge test followed by GPs and paramedics. In contrast, physicians, surgeons, pediatricians and medical students had comparable scores. No candidate performed all CPR steps correctly in the proper sequence. In addition, the manikin performance of all groups was poor. Nevertheless, the course significantly improved the theoretical knowledge (P < 0.0001) and performance in both CPR steps (P < 0.0005) and manikin performance (P < 0.0005). The multivariate discriminant analysis identified that experience and prior CPR training influenced significantly (P < 0.001) the degree of retention of theoretical knowledge only, but not the actual performance of basic resuscitation. It is concluded that no assumption based on previous clinical knowledge should be made for the expected CPR performance of all doctors, regardless of speciality. Formal training programs in medical schools should be considered. Our data also indicate that training is the only objective way to improve performance of all the candidates including the medical students.

Education, medical staff, medical students - Complications, cardiac arrest - Heart, cardiac massage - Cardiopulmonary resuscitation.

INTRODUCTION

In teaching hospitals the first respondent to cardiac arrest code are interns and residents. Although they have detailed knowledge of pathophysiology, most of them do not know cardiopulmonary resuscitation (CPR). Several reports have demonstrated that the overall performance of basic cardiac life support (BCLS) by physicians is poor [1,21. In fact, in one study, the performance of physicians and registered nurses was found to be comparable to that of the general public [3].

*Presented in part in abstract form at the 6th World Congress on Emergency and Disaster Medicine, 1989, Hong Kong.

**Author to whom reprint requests should be sent.

In this study we examined the performance of different medical staff and medical students in BCLS in a simulated cardiac arrest situation.

MATERIALS AND METHODS

Two hundred and twenty-four candidates (31 anesthesiologists, 52 general practitioners (GPs), 33 physicians, 41 surgeons, 4 cardiologists, 7 pediatricians, 32 medical students and 24 paramedical personnel) attending a BCLS course were tested in their theoretical knowledge 'knowledge test' of BCLS. The test consisted of 15 multiple choice written questions (MCQS) with one correct answer per question. The same MCQs were included in the post-test but with different sequence. The post-test consisted of 50 written MCQs and was conducted at the end of the course. We calculated the number of incorrect answers for each candidate in both pre- and post-test out of the 15 MCQS.

We also evaluated the skill of each candidate before and after the BCLS course according to American Heart Association (AHA) criteria during performance of single-rescuer CPR on a recording manikin. We used a check list (Table 1) and recording strip from each manikin (Fig. 1) for evaluation of CPR steps and manikin performance respectively. The 'CPR steps' has been organized into three categories: assessment and calls for assistance, ventilation and compression in the check list, and a score ranging from 0 to 5 was assigned to each step as depicted in Table I with a maximum score of 80. The time factor was taken into consideration when evaluating CPR steps. A score of 20 has been assigned for the correct 'manikin performance' as shown in Fig.1 The scoring system employed in this study allowed us to evaluate similar steps in different categories and to make generalization about performance in each category.

Statistical analyses were carried out with KruskaI-Wallis one-way analysis of variance, Mann-Whitney and sign tests where applicable using BMDP (1988) statistical package.

In order to predict which factors might contribute to the retention of theoretical knowledge and manikin performance skills at the end of the course, we used a multivariate discriminant analysis technique. Independent variables included experience, prior CPR training, sex and pre-course score in the theoretical knowledge. The same procedure was repeated using the pre-course scores in CPR steps and manikin performance instead of the score in the theoretical knowledge respectively. All independent variables were regressed on the perspective score at the end of the course.

RESULTS

Results are expressed as Medians ± standard error of median. The distribution of the candidates according to sex, status, number of prior CPR courses attended and experience is shown in Table II.

Table 1. Check list.

Score Required

Performance

time (s)

Assessment of responsiveness and calls for help 4-10

- Sound 5 4 3 2 1 0

- Touch 5 4 3 2 1 0

- Stimulus 5 4 3 2 1 0

- Performance 5 4 3 2 1 0

- Call for help 5 4 3 2 1 0

- Airway opening 5 4 3 2 1 0 2

- Look, listen and feel 5 4 3 2 1 0 3-5

Breathing

- Twice 5 4 3 2 1 0 4

- Slowly 5 4 3 2 1 0 4

- Volume 5 4 3 2 1 0

- Mouth to mouth

- Mouth to nose

- Carotid pulse/IF no pulse activate EMS 5 4 3 2 1 0 5-10

- Location 5 4 3 2 0 1

- Compression/relaxation (15 compressions) 5 4 3 2 1 0 9-11

- Compression/breathing ratio (I 5 : 2) 5 4 3 2 1 0

- 4 Cylces of 15 : 2 5 4 3 2 1 0 52-73

- Check up pulse and ventilation/ if

absent activate EMS 5 4 3 2 1 0 5

IINTERVAL 5-7 S.

PAUSE TIME 9-11 S. 5 POINTS

2 POINTS 2 POINTS

MAXIMUM SCORE = 20

Fig. 1. Scoring of the manikin performance. Ten points assigned to manikin compression (Details are shown in the figure). Five points assigned to ventilation and five points for allowing 5-7 s between compression and ventilation.

The results of the theoretical knowledge test before and after the course revealed a significant improvement (P < 0.0001) with the training in all groups except for the cardiologists. The latter group performed well in the pre-test, therefore, no significant improvement was noted in the post-test. Anesthesiologists and cardiologists demonstrated the best performance followed by GPs and paramedical personnel. Surprisingly, physicians, surgeons, pediatricians and medical students had a comparable performance. The latter groups had an overall poor performance (Table 11).

Table 11. Sex, status, experience and number of CPR courses attended. Median (S.E. median).

Paramedics Anaesthe- GPs. Physicians Surgeons Cardiolo- Pediatrians Medical

siologists gists students

(n = 24) (n = 31) (n = 52) (n = 33) (n = 4) (n = 4) (n = 7) (n = 32)

Sex (M/F) 8/16 26/5 36/16 24/9 27/14 4/0 7/0 0/32

Status

Senior 7 8 3 7 13 0 5 -

Junior 17 23 49 26 28 4 2 -

No. of CPR 1 (0) 1 (0) 1 (0) 1 (0) 1 (0) 1 (0) 1 (0) 1 (0)

courses

attended

Prior 6/18 21/10 17/35 16/17 13/28 4/0 6/1 7/25

experience

in CPR Yes/No

Table Ill. Results of knowledge test. The number of incorrect answers are shown. Median (S.E. Median).

Paramedics Anaesthe- GPs. Physicians Surgeons Cardiol- Pediatrians Medical

(n = 24) (n = 3 1) (n = 52) (n = 33) (n = 41) (n = 4) (n = 7) (n = 32)

Score 5 (0.86)*t 2 (0.8) 4 (0.57) 6.0 (0.28)*t 7 (0.86)*,**,t 1.5 (1.15) 8 (1.15)*,**-t 7.5 (0.57)*,**-t

Before the

course

Score § 3 (0.8)*,**,t 0 (0) 0 (0.57)* 3 (0.86)*,t 2 (0.57)*-t 0 (0) 3 (1.44)*,t 2 (0.57)*t

after the

course

*P = 0.0 I or less when compared with Anesthesiologists.

The student groups had the poorest performance (P = 0.0006 or less) when compared to other groups (Table IV).

Training significantly (P < 0.0005) improved performance in all groups. There were no significant differences in the scores among the groups after the course.

The overall performance of all groups before the course was poor (Table V). Training significantly (P < 0.0005) improved performance in all groups. There were no significant differences in the scores among the groups after the course.

Multivariate discriminant analysis indicated that the degree of retention of theoretical knowledge at the post-test was related to experience (P < 0. 00 1) and to prior life support training (P < 0.001). No other factors or recognizable combination of factors could provide an additional statistically significant increment in classification power. On the other hand, we could not identify any significant factor that had statistically significant independent correlation with the scores after the course with either CPR steps or manikin performance.

DISCUSSION

Different studies have demonstrated that CPR is attempted for 1-2% of all hospital admissions and 30% of all patients who die [4,51. It was surprising that the resuscitation success rates were low. When CPR was attempted in the hospital, only 5-20% of patients survived and returned home [4-10]. Although many patientrelated factors have been proposed [4], the low rate of success of in-hospital resuscitation could be also attributed to lack of formal life-support training of the physician-bystanders. In fact, several reports have indicated that many physicians do not know CPR [ 1, 1 1 ].

In this study we examined the theoretical knowledge and skills of different medical profession personnel in BCLS. Anesthesiologists and cardiologists had the best theoretical knowledge before the course, followed by physicians and paramedics. On the other hand, surgeons, pediatricians and medical students had the least knowledge in BCLS. The course significantly (P < 0.0001) improved the theoretical knowledge for all groups except for the cardiologists as they scored highly in the pretest.

The scoring system employed in this study permitted an objective evaluation of the performance of each category with respect to CPR steps and manikin performance. The overall performance of all groups was very poor before the course, both in the sequence of CPR and in manikin performance. Although Kouwenhoven et al. [12] stated nearly 30 years ago that ' ... all that is needed are two hands', it has never been demonstrated that the medical staff can perform CPR correctly despite their medical training and experience. In addition, although the anesthesiologists scored

Table IV. Scores in CPR steps. Median (S.E. median). The maximum score for correctly performed steps is 80.

Paramedics Anaesthe- GPs. Physicians Surgeons Cardiolo- Pediatrians Medical

siologists gists students

Score before 7 (5.1) 53.3 (8.9)* 10.5 (3.7) 4 (12.1) 3 (10) 5 (4) 5 (2.3) 0 (0)

the course

Score ** after 78 (1.1) 79 (0.8) 76 (1.4) 78 (1.7) 75 (1.6) 78 (1.5) 78 (1.7) 76.5 (8.6)

the course

*Different from other groups (P < 0.01).

**Training significantly (P = 0.0006 or less) improved performance in all groups.

Table V. Manikin Performance. Median (S.E. median). The maximum score for correct performance is 20.

Paramedics Anaesthe- GPs Physicians Surgeons Cardiolo- Pediatrics Medical

siologists gists students

Score before 4 (2.3) 7.5 (1.7) 2.5 (2.3) 0 (2.3) 0 (0) 3 (2.8) 3 (2.3) 0 (0)

the course

Score after * I 19 (I. 1) 18 (0.8) 17 (0.5) 18 (1.4) 17 (1.7) 18 (0.9) 17 (0.8) 17 (0.5)

the course

*Training significantly (P < 0.0005) improved performance in all groups.

better in the CPR steps, no one performed all steps correctly in proper sequence, the overall performance of other groups of medical profession in CPR steps and manikin performance was comparable to that of the medical students. When Webb and Lambrew [111 tested 35 physicians in basic resuscitation tasks (airway, breathing, circulation) only 22% were able to compress and ventilate the manikin adequately in a simulated cardiac arrest. Similarly, Lowenstein et al. [1] tested 45 medical and surgical house officers in both basic and advanced resuscitation skills. Only 29% could properly compress and ventilate the manikin. In addition, Kay and Mancini [3] found that no medical resident or registered nurse and only one layperson performed each step correctly, and in proper sequence.

The result of the discriminant analysis indicated that experience and prior CPR training influenced the degree of retention of theoretical knowledge only but not the actual CPR performance in proper sequence. However, we found that training significantly (P < 0. 0005) improved performance of all groups including the medical students in both CPR steps and manikin performance. The implication of this is that the superior CPR skills could be achieved through training of CPR.

It is mandatory for the new residents in our department to hold both BCLS and ACLS certificates during the first 3 months of training. In addition, the curriculum for the undergraduates has been modified to include a proper training in BCLS and Modified Advanced Cardiac Life Support. Lowenstein et al. [131 showed that survival after in-hospital cardiac arrest was significantly increased if house officers who staff the code teams were trained in basic and advanced cardiac life support.

CONCLUSIONS

Our data demonstrate that all doctors, regardless of speciality, can not perform BCLS in proper sequence. The performance of medical students in a simulated cardiac arrest situation was comparable to that of other medical professions, therefore, no assumptions should be made for any previous theoretical or clinical knowledge. This stresses the need for a formal training program in medical school. Our data also indicate that prior training in CPR might have an influence on the degree of retention of theoretical knowledge but not on the performance of basic resuscitation. Regular training in BCLS is required to attain CPR skills.

ACKNOWLEDGEMENTS

We thank Dr. Roel Ilada for his assistance and Ms. Melba T. Atienza for her expert secretarial help.

REFERENCES

6 J.G. Lemire and A.L. Johnson, Is cardiac resuscitation worthwhile? A decade of experience, N. Engl. J. Med., 286 (1972) 970-972.

7 R.C. Peatfield, D. Taylor, R.W. Stillett and M.W. McNicol, Survival after cardiac arrest in hospital, Lancet, 1 (1977) 1223-1225.

8 B. Messert and C.E. Quarliere, Cardiopulmonary resuscitation: perspective and problems, Lancet, 11 (1976) 410-411.

9 C.O. Hershy and L. Fisher, Why outcome of cardiopulmonary resuscitation in general wards is poor, Lancet, 1 (1982) 31-34.

13 S.R. Lowenstein, E.M. Sabyan, C.F. Lassen and D.C. Kern, Benefits of training physicians in advanced cardiac life support, Chest, 89

(1986)512-516.