Resuscitation 85 (2014) 472–479

Contents lists available at ScienceDirect

Resuscitation journal homepage: www.elsevier.com/locate/resuscitation

Review article

Incidence and outcome from adult cardiac arrest occurring in the intensive care unit: A systematic review of the literature Ilmar Efendijev a,∗ , Jouni Nurmi a , Maaret Castrén b,c , Markus B. Skrifvars a a

Department of Anesthesiology and Intensive Care Medicine, Helsinki University Central Hospital, Finland Karolinska Institutet, Department of Clinical Science and Education, Södersjukhuset and Section of Emergency Medicine, Södersjukhuset, Stockholm, Sweden c Department of Emergency Medicine, University of Turku, Finland b

a r t i c l e

i n f o

Article history: Received 6 October 2013 Received in revised form 19 December 2013 Accepted 20 December 2013 Keywords: Intensive care unit Cardiac arrest Resuscitation

a b s t r a c t Background: Significant amount of data on the incidence and outcome of out-of-hospital and in-hospital cardiac arrest have been published. Cardiac arrest occurring in the intensive care unit has received less attention. Aims: To evaluate and summarize current knowledge of intensive care unit cardiac arrest including quality of data, and results focusing on incidence and patient outcome. Sources and methods: We conducted a literature search of the PubMed, CINAHL and Cochrane databases with the following search terms (medical subheadings): heart arrest AND intensive care unit OR critical care OR critical care nursing OR monitored bed OR monitored ward OR monitored patient. We included articles published from the 1st of January 1990 till 31st of December 2012. After exclusion of all duplicates and irrelevant articles we evaluated quality of studies using a predefined quality assessment score and summarized outcome data. Results: The initial search yielded 794 articles of which 780 were excluded. Three papers were added after a manual search of the eligible studies’ references. One paper was identified manually from the literature published after our initial search was completed, thus the final sample consisted of 18 papers. Of the studies included thirteen were retrospective, two based on prospective registries and three were focused prospective studies. All except two studies were from a single institution. Six studies reported the incidence of intensive care unit cardiac arrest, which varied from 5.6 to 78.1 cardiac arrests per 1000 intensive care unit admissions. The most frequently reported initial cardiac arrest rhythms were nonshockable. Patient outcome was variable with survival to hospital discharge being in the range of 0–79% and long-term survival ranging from 1 to 69%. Nine studies reported neurological status of survivors, which was mostly favorable, either no neurological sequelae or cerebral performance score mostly of 1–2. Studies focusing on post cardiac surgery patients reported the best long-term survival rates of 45–69%. Conclusions: At present data on intensive care unit cardiac arrest is quite limited and originates mostly from retrospective single center studies. The quality of data overall seems to be poor and thus focused prospective multi-center studies are needed. © 2014 Elsevier Ireland Ltd. All rights reserved.

Contents 1. 2.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Sources and methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Data extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4. Quality score . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5. Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6. Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

473 473 473 473 473 473 474 474

∗ Corresponding author at: Department of Anesthesiology and Intensive Care Medicine, Helsinki University Central Hospital, Topeliuksenkatu 5, PL 266, 00029 HUS, Helsinki, Finland. E-mail addresses: [email protected], ilmar.efendijev@hus.fi (I. Efendijev). 0300-9572/$ – see front matter © 2014 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.resuscitation.2013.12.027

3.

4. 5.

I. Efendijev et al. / Resuscitation 85 (2014) 472–479

473

Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Study type, size and time of publication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Description of included ICUs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Patient profile and incidence of ICUCA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Definition of cardiac arrest and return of spontaneous circulation (ROSC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5. Etiology and initial cardiac rhythm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6. Survival and outcome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.7. Data on DNAR/DNR policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conflict of interest statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

475 475 475 475 477 477 477 478 478 478 479 479 479 479

1. Introduction Cardiac arrest (CA) is a major health problem and survival from cardiac arrest depends largely on prompt initiation of cardiopulmonary resuscitation. Cardiac arrest occurring in the intensive care unit (ICU) represents a specific sub-group, which differs in several ways from cardiac arrests occurring outside the hospital or in other areas of the hospital. Vital functions of patients treated in the intensive care unit are monitored continuously and the risk of unwitnessed arrests is low.1,2 The delay to initiation of cardiopulmonary resuscitation (CPR) is likely to be short which should result in better chances for survival.1,2 Advanced resuscitation, including invasive treatment options and differential diagnosis of cause of cardiac arrest are readily available in the ICU. On the other hand pre-existing conditions and severity of illness are likely to have a negative impact on survival. Data on out-of-hospital (OHCA) and in-hospital cardiac arrest (IHCA) are accumulating but data on incidence and outcome of intensive care unit cardiac arrest (ICUCA) are scarce, and appear to be of a highly variable quality. The aim of the present study was to systematically review the literature on ICUCA with focus on demographics, incidence and outcome of ICUCA. We sought to evaluate the quality of the included studies with pre-defined quality criteria. 2. Methods 2.1. Sources and methods We performed a literature search in January 2013 using PubMed, CINAHL and Cochrane Database of Systematic Reviews (1.1.1990–31.12.2012), using the following medical subheadings (MeSH): “heart arrest” AND “intensive care unit” OR “critical care” OR “critical care nursing” OR “monitored bed” OR “monitored ward” OR “monitored patient”. In addition we manually searched all references of selected articles for cross-references. In the process of selection of literature and data extraction PRISMA statement and PRISMA checklist were followed.3 2.2. Selection Following the initial search three authors independently (JNU, MSK, MCA) screened articles for further analysis based on the article title and/or abstract when available. After that one author (IEF) reviewed full texts of selected articles and performed an exclusion of duplicates and studies meeting our exclusion criteria. We excluded studies conducted in the pediatric population, studies limited to reporting CA occurring in coronary care or high dependency units and studies that focused primarily on other subjects than ICUCA patients such as studies focusing on ICU care of

out-of-hospital and in-hospital cardiac arrest patients. Inclusion criteria for the studies were: studies focused on ICUCA, adult population, studies reporting data on ICUCA outcome and/or incidence. 2.3. Data extraction We reviewed all eligible articles independently in order to extract all relevant data, such as year of publishing, type of study, study setting, patient profile, ICU characteristics, cardiac arrest etiology, duration of resuscitative efforts, incidence, mortality, neurologic and long-term outcomes. We made no attempt to contact the authors in order to confirm or to obtain additional data. 2.4. Quality score We developed a quality assessment score (Table 1) in order to evaluate methodological quality of the selected articles. According to the scoring table a score of 0 would indicate poor quality and Table 1 Evaluation criteria for methodological quality of the included studies. Evaluation criteria

Points, max 22

1. Study type Focused prospective study (4 points) Prospective resuscitation registry (3 points) Prospective ICU registry (2 points) Retrospective (1 point) 2. ICU definition/staffing reported 3. General ICU/patient profile outlined 4. Cardiac arrest incidence: cardiac arrest/ICU admission 5. Cardiac arrest clearly defined 6. Initial arrest rhythm reported 7. Cardiac arrest etiology reported 8. Initial ICU admission diagnosis 9. Time intervals (1 point each) Time to initiation of CPR Time to ROSC 10. Reporting of clinical factors found to correlate with the occurrence of cardiac arrest (e.g. changes in physiological parameters, acute illness scoring systems, etc.) 11. Outcome data (1 point each) 24 h survival ICU discharge Hospital discharge Long-term survival (minimum 180 days) 12. Neurological outcome (1 point) or quality of life (1 point) after hospital discharge 13. Data on special resuscitation techniques and unconventional treatment strategies (e.g. open-chest CPR)

4

1 1 2 1 1 1 1 2

1

4

2 1

ICU, intensive care unit; CPR, cardiopulmonary resuscitation; ROSC, return of spontaneous circulation.

474

I. Efendijev et al. / Resuscitation 85 (2014) 472–479

Table 2 Summary of selected studies based on study type. Study

Study type

Setting

Years

N

Included patients

Cardiac arrest incidence

Tortolani et al. 1990

Retrospective

US tertiary hospital

NR

470 IHCA/158 ICUCA

NR

6

Peterson et al. 1991 Landry et al. 1992 Karetzky et al. 1995 Smith et al. 1995 Anthi et al. 1998 Wallace et al. 2002 Myrianthefs 2003 Rabinstein et al. 2004

Retrospective

US university hospital/ US university hospital US tertiary hospital

1985–1988

114

All IHCA grouped by CA location MICU

NR

8

1987–1988

114

MICU and SICU

NR

9

1990–1992

IHCA

NR

6

1987–1993

668 IHCA/360 ICUCA 55

SICU

55/5237

10

1993–1996

29

CICU

29/3982

16

1993–2000

406

MICU

406/5196

10

1999–2000

111

General ICU

NR

7

1994–2001

NICU

NR

10

1999–2003

21 ICUCA/38 (incl. IHCA admitted to NICU) 169

Mixed ICU

169/16 898

14

1992–2002

214

NICU

NR

10

2004–2006

222

ICU

NR

14

1998–2004

148

CICU

NR

9

2002–2007

83

ICU

NR

7

2000–2008

49 656

ICU

NR

8

2000–2005

ICU + CCU + CICU

NR

9

2008–2010

239 ICUCA/517 (ICU + CCU) 22

22/3931

14

2009–2010

131

ICUCA (+IHCA + OHCA) ICU

131/9975

10

Retrospective Retrospective Prospective ICU registry Focused prospective study Retrospective Retrospective Retrospective

Enohumah et al. 2006 Yi et al. 2006 Chang et al. 2009 Guney et al. 2009

Retrospective

Grigoriyan et al. 2009 Tian et al. 2010

Retrospective

Kutsogiannis et al. 2011 Skrifvars et al. 2012 Lee et al. 2013

Retrospective Focused prospective study Retrospective

Prospective resuscitation registry Retrospective Focused prospective study Retrospective

US university hospital Greece tertiary teaching hospital US university hospital Greece general hospital US tertiary hospital

Germany university hospital South Korea university hospital Taiwan university hospital Thoracic and cardiovascular center, Turkey US community teaching hospital 400 facilities, US

Multicenter, Canada Australia tertiary hospital South Korea university hospital

Quality score

NICU, neurointensive care unit; CICU, cardiac intensive care unit; SICU, surgical intensive care unit; MICU, medical intensive care unit; CCU, coronary care unit; NR, not reported.

a score of 22 would indicate excellent quality. We scored studies based on study type, i.e. retrospective, prospective registry based or focused prospective study. Further on we scored studies based whether the manuscript included data on the following: (1) Characteristics of the included ICUs (level of care intensity, patient profile or specialty and staffing/ICU resources), (2) characteristics of the baseline patient population treated in the ICU (admission diagnosis), (3) cardiac arrest management related data (ROSC = return of spontaneous circulation/duration of CPR = cardiopulmonary resuscitation, initial cardiac rhythm, definition of the cardiac arrest as lack of clinical signs of circulatory function), (4) outcome (survival to hospital discharge and long-term survival, as well as neurological and/or functional status of the survivors, when available), and (5) data on special resuscitation techniques used, e.g. open-heart massage, extracorporeal support of circulation and/or oxygenation.

2.5. Definitions We defined cardiac arrest as lack of clinical signs of the circulatory function (e.g. absence of palpable pulse), unresponsiveness and/or apnea, agonal breathing (if not on artificial ventilation), which lead to the initiation of CPR.4 Outcome data was evaluated according to the duration of survival of the patients: initial survival defined as survival for 24 h after ICUCA event, survival

to ICU discharge, survival to hospital discharge, long-term survival defined as survival over six months or longer follow-up period. We defined good quality of long-term survival as ability of survivors to complete everyday activities independently or with minor help and good neurological outcome according to cerebral performance category (CPC) of 1–2 or good performance and moderate disability, respectively. Unfavorable outcome would mean post ICUCA patient suffering from severe medical conditions or functional disabilities that affect every day life and unfavorable neurological outcome according to cerebral performance category of 3–4 or severe disability and coma/vegetative state, respectively.5 Incidence was calculated according to American Heart Association’s recommendations for IHCA as total number of ICUCA patients divided by the number of patients admitted to the ICU, all known do-not attempt resuscitation (DNAR)/do-not resuscitate (DNR) patients were excluded.6

2.6. Evaluation Two authors (IEF, MSK) independently evaluated the selected articles’ methodological quality using the adopted assessment score (Table 1). The weighted kappa score was used to assess agreement between the two reviewers with fair result (0.339).

I. Efendijev et al. / Resuscitation 85 (2014) 472–479

475

Fig. 1. Flow chart of the selection process.

Disagreement between authors was solved with an independent review of a third author (JNU) after which consensus was achieved. 3. Results The search strategy initially identified 794 citations (Fig. 1). Three articles were added via search for cross-references. One article was identified manually; it was published electronically in March 2013 after initial database search was completed. After assessment of the articles for stated inclusion and exclusion criteria based on title, abstract and full text the selection process resulted in 19 articles. One article was excluded because of significant similarity7 with another included study8 by the same author, thus giving us a number of 18 eligible studies for the review.3 3.1. Study type, size and time of publication Of eighteen included studies there were five prospective studies (Table 1). Three of those were prospective studies focused on ICUCA, one prospective ICU registry study and one prospective resuscitation registry study. Rest of the studies were retrospective. Eight

studies were published between 2000 and 2010, six of the articles dated back to 1990s and four articles were published after 2010. One prospective resuscitation registry study was conducted at 400 different hospitals in the US as a part of the “Get with the Guidelines Registry”.9 One other study was a prospective multi-center study and the rest originated from single institutions (Table 2).10 3.2. Description of included ICUs Twelve out of 18 studies did not give a clear ICU description nor did they report ICU staffing. Seven studies reported the number of ICU beds, which ranged from 8 to over 200 with a mean of 68 and median of 42 beds.1,2,8,10–13 Only one study reported data on nurse to patient ratios,10 and nine studies reported data on level of experience of the on-site physician, who was mainly of resident or senior resident level.2,8,11,12,14–18 3.3. Patient profile and incidence of ICUCA Patient profile according to medical specialty and/or type of disease treated was clearly specified in ten studies. Of those 10

476

Table 3 Summary of selected studies based on patient profile. Patient profile

CA etiology

Duration of CPR or time to ROSC

Survival to hospital discharge

Long-term survival/duration of follow up (in months)

Neurological outcome

Comment

Tortolani et al. 1990

NR

NR

Duration of CPR in all ICUCA patients 38.6 ± 26.3 min

14%

NR

NR

Peterson et al. 1991

NR

NR

11%

5.3%/42 months

NR

Landry et al. 1992

NR

NR

Time to ROSC in survivors to hospital discharge 8 ± 4 min Length of code in survivors to hospital discharge 12 ± 10 min

ICUCA and IHCA data mixed, outcome distinguished by CA location CA and RA are included in the same group

5%

1.8%/12 months

CA and RA are evaluated as the same group

Smith et al. 1995

Surgical ICU patients

NR

13%

NR

Karetzky et al. 1995 Anthi et al. 1998

NR

NR

Duration of CPR in survivors (mean) 14.43 min NR

LTS n = 2: n = 1 quadriplegia, n = 1 chronic medical problems NR

3.3%

2.5%/12 months

NR

Cardiac surgery patients

MI, tamponade, graft malfunction, unknown

NR

79%

69%/12 months

No neurological sequelae in survivors at discharge

Cancer patients (leukemia, solid tumor, lymphoma/myeloma, HSCT recipient, 2 with no malignancy) Mixed population based on diagnosis on admission

NR

NR

2%

1.2%/over 6 months

NR

CA incidence for hospital admissions Patients on maximal inotropic support, IABP and VAD patients were excluded 2 patients lost to follow up

NR

NR

0%

Wallace et al. 2002

Myrianthefs 2003

NR

Rabinstein et al. 2004

NICU

Detailed report

Duration of resuscitation efforts in survivors ≤5 min

18%

Yi et al. 2006

NICU

NR

Time to ROSC in survivors 8.3 ± 4.87 min

8.9%

18% (1–24 months/mean 9 months follow up) 6.5%/12 months

Enohumah et al. 2006

Mixed: includes CICU and NICU

Reported (Utstein) Majority presumed cardiac

Duration of CPR in survivors (mean) 10.3 min

47.3%

NR

Chang et al. 2009

Mixed (MICU, NICU, SICU, CCU)

NR

Time to ROSC in survivors (mean) 7 min

15.3%

10.9%/12 months

Grigoriyan et al. 2009

Mixed population

NR

NR

17%

NR

Guney et al. 2009

CABG patients

PMI 70%

NR

over 60%

45%/37 ± 25 months

5/7 survivors Rankin scale score ≤232 At one year follow up: 1 unconscious 13 conscious CPC-score: % of survivors to hospital discharge 1–2: 82.5% 3–4: 17.5% CPC-score: % of survivors to hospital discharge 1–2: 77.4% 3–4: 9.7% Data missing 12.9% NR

NR

Immediate survival 100%, 24 h survival 9.2% Patients’ age 16–81 years

Out of 188 ICUCA 19 were excluded (2 DNAR patients, in 17 patients CPR was not attempted because it was considered futile) 4% of all patients lost to follow up

CA and RA are evaluated as the same group

I. Efendijev et al. / Resuscitation 85 (2014) 472–479

Study

NR NR One month survival: 24.4% Three months survival: 20.6% Resp. failure n = 52 Card. failure n = 45 Hypovolemic shock n = 34 NR Lee et al. 2013

LTS, long-term survival; CPC, cerebral performance category; CABG, coronary artery bypass; CA, cardiac arrest; RA, respiratory arrest; IHCA, in-hospital cardiac arrest; OHCA, out-of-hospital cardiac arrest; MI, myocardial infarction; PMI, perioperative myocardial infarction; IABP, intra-aortic balloon pump; VAD, ventricular assist devices.

CPC 1–2 95.6% of all survivors (OHCA, IHCA; ICUCA) at hospital discharge CPC: % of survivors at three months CPC 1–2: 48% CPC 3–4: 52% NR 52% NR NR Skrifvars et al. 2012

Time to ROSC in survivors (3 months) 9 ± 13 min; in all patients 18 ± 20 min

NR 26.9% NR NR Kutsogiannis et al. 2011

Duration of CPR (all patients) mean 19.9 min median 14 min Time to ROSC in survivors Median 5 min

24.3%/12 months 15.9%/60 months

Two groups: vasopressor and non-vasopressor. Survival more favorable in the non-vasopressor group Survival data CCU + ICU CPC-score % of survivors to hospital discharge 1–2: 82.7% NR 15.9% Reported NR Tian et al. 2010

NR

Neurological outcome CA etiology Patient profile Study

Table 3 (Continued)

Duration of CPR or time to ROSC

Survival to hospital discharge

Long-term survival/duration of follow up (in months)

Comment

I. Efendijev et al. / Resuscitation 85 (2014) 472–479

477

Table 4 Comparison of ICUCA incidence. Study

Study period (months)

Incidence

Prevalence

Incidence CA/1000 admissions

Smith et al. 1995 Anthi et al. 1998 Wallace et al. 2002 Enohumah et al. 2006 Skrifvars et al. 2012 Lee et al. 2013

66 30 96 60

55/5237 29/3982 406/5196 169/16 898

1.1% 0.7% 7.8% 1%

10.5 7.3 78.1 10

24 18

22/3931 131/9975

0.6% 1.3%

5.6 13.1

studies four were from mixed-population ICUs, four were from specialized units: two cardiac surgery ICUs and two neurosurgery ICUs. One study originated from a cancer center ICU treating mixed non-cardiac surgical patients and one study was from non-cardiac surgical ICU (Table 3). The prevalence of the ICUCA varied from 0.6% up to 7.8%18,19 and the incidence of ICUCA, calculated as number of cardiac arrests per 1000 ICU admissions was between 5.6 and 78.1 (Table 4).13,17 3.4. Definition of cardiac arrest and return of spontaneous circulation (ROSC) Three of the studies1,13,20 did not report any definition of cardiac arrest and three studies12,15,16 included both cardiac and respiratory arrests. Uniformly cardiac arrest was defined as cessation of circulatory function evidenced by the lack of a palpable pulse resulting in the initiation of a CPR algorithm. Time to ROSC was reported in eleven studies. 2,8,10,11,13,15–17,19,21,22 Reported time to ROSC for survivors to hospital discharge varied from less than 5 min up to 65 min (Table 3).17,22 Two studies reported time to initiation of CPR, which was less than 30 s.1,2 3.5. Etiology and initial cardiac rhythm Less than half of the studies reported the presumed etiology of the cardiac arrest, and slightly different definitions were used (Table 3). Two studies on cardiac surgery patients reported etiology of CA as being related to an acute cardiac event, including myocardial infarction (either independent or due to graft malfunction) and tamponade.20,23 As reported by Rabinstein and colleagues, most of the ICUCA events in a neurologic-neurosurgical ICU occurred due to noncardiac reasons, the study included specific CA etiology data. In the study by Enohumah et al., most of the ICUCAs were due to a presumable cardiac cause. The largest study included in this review reported acute myocardial infarction as the immediate reason of CA in only 9% of the cases, most commonly (in 19% of the patients) the cause of CA was septicemia.9 In the latest study by Lee and colleagues cardiac failure accounted for 26% of all cardiac arrests including CPR on arrival13 (Table 3). The reporting of initial cardiac arrest rhythm was variable; thirteen studies reported the distribution of initial cardiac rhythms.1,2,8–14,16,17,19,22,23 Most common cardiac rhythms of ICUCA patients were non-shockable rhythms with rates varying from 55% to 84%. 3.6. Survival and outcome Initial survival (survival over 24 h) rates were reported in most studies1,2,8,10,12,15,18,21 and ranged from 9% up to over 90%.1,20 Hospital discharge rates varied from 0 to 79%.1,23 Two studies that

478

I. Efendijev et al. / Resuscitation 85 (2014) 472–479

included only cardiac surgery patients reported survival to hospital discharge rates from 60% to 79%, although in latter study patients on maximum inotropic and/or intra-aortic balloon pump support, as well as patients with ventricular assistant devices were excluded.20,23 Survival of ICUCA in units treating neurosurgical patients was between 9% and 18%.2,22 Yi and colleagues have noted that in a neurointensive care unit population none of the patients with initial infectious conditions (septicemia and pneumonia) survived to hospital discharge. One study conducted in the ICU of a cancer center reported a survival rate to hospital discharge of 2%.18 Nine studies reported data on long-term survival. The highest survival rates of 45% and 69% during a one year follow up period originated from units treating cardiac surgery patients.20,23 Wallace and colleagues reported the lowest survival rate of 1.2% at 6 months follow-up in patients diagnosed with a malignant disease. Five other studies from mixed population ICUs reported long-term survival rates at one-year follow-up in between 1.8% and 24% (Table 3). Nine studies included data on the neurological and/or functional outcome of survivors (Table 3). Yi et al. reported unfavorable outcome in the NICU population with only 5% achieving complete recovery of all ICU patients who received CPR. In the study by Landry and colleagues of two long-term survivors one had quadriplegia and one had several chronic medical problems that were not specified in the paper. Rest of the studies reported favorable neurological outcome. Anthi et al. reported most optimistic results with a 69% survival rate and no neurological sequelae among survivors. In the latest study by Lee et al., 48% of the survivors had CPC 1–2 and 52% CPC 3–4. 3.7. Data on DNAR/DNR policies Eleven studies included mention of DNAR/DNR orders. 1,2,8,11,13,15,17–19,21,22 Of those one included a description of practical

implications, but no mention of decision criteria.18 Myrianthefs and colleagues had described the policy of their institution as refraining from resuscitative actions in ICUCA patients who are already being on maximum vasopressor and inotropic support with underlying acute or chronic untreatable disease. Two studies had reported that their institutions had no official DNAR/DNR policies at the time of data collection.8,17 The paper by Chang et al. had the most extensive explanation of DNAR, including obtaining consent and its practical implications, although there was no mention in the paper whether this was an official policy of the institution.11 4. Discussion According to our review current data on ICUCA is mainly based on single center retrospective studies with very variable study populations. Methodological quality is of modest level according to the quality score developed specifically for this review. Survival is highly variable and differs based on study setting. The highest survival seems to be in cardiac surgery patients. Data on long-term outcome, cause of death and quality of life of survivors is limited. The great variability in reported survival rates is probably related to the heterogeneity of study populations.1,23 The best prognosis seems to be in post cardiac surgery patients, in whom the cause of cardiac arrest was often related to either myocardial infarction due to graft malfunction and surgical complications such as tamponade or ventricular tachyarrhythmias.20,23 On the other hand patients from mixed ICUs probably have multiple factors contributing to the occurrence of the cardiac arrest and this might indicate complete physiological failure and thus aggressive and well-timed CPR will be of limited value. One can also speculate whether the marked variability in ICUCA survival might in part be related to differences

in resuscitation and DNAR policies at different institutions as well as to differences in ICU admission criteria. Interestingly, by far the highest incidence of ICUCA was reported in a cancer center.18 A factor that makes comparison difficult is that oldest studies did not use the “in-hospital Utstein standard”, published in 1997.4 It also remains unclear whether implementation of hospital-wide rapid response teams/medical emergency teams might increase ICUCA incidence by transferring potential IHCA event to the ICU area.24,25 As expected the initiation of CPR in case of ICUCA was fast and reported to be less than 30 s. Duration of resuscitation efforts in the survivors was mostly under 10 min, and time seems to be an important factor affecting survival in the ICUCA population, although prolonged resuscitation did not necessarily rule out the possibility of positive outcome. Two studies concluded that ICU patients who experienced cardiac arrest have better survival, if they were not on vasopressors at the moment of ICUCA or immediately previously to that.9,12 This is likely to indicate circulatory shock and clearly decreases the likelihood of a good outcome. Most of the ICUCA events’ initial cardiac rhythm was reported to be non-shockable which is expected given the low likelihood of sudden cardiac ischemia in the ICU CA population. The result is similar to IHCA cases were VT/VF is the initial rhythm in about 25–30% which is less than in OHCA.26–31 Half of the studies reported long-term survival rates and limited data on the quality of life of ICUCA survivors. Only one study reported data on the causes of death of long-term survivors.2 Thus for the majority of the studies it is still unknown whether cause of death of long-term survivors was related to sequelae from the ICUCA such as neurologic injury, the disease that initiated intensive care or other comorbidities. We were unable to identify any study that compared ICUCA patients with rest of the ICU population in order to identify factors that might predict to some extent the ICUCA event or correlate with it. As we faced great variety in the outcomes of the patients it is impossible to make any conclusion whether all ICU patients, except for those with “do not resuscitate” decision, should be resuscitated in the event of cardiac arrest. It would be interesting to see the results of the prospective study focusing specifically on the ICUCA population the conduction of which will need several centers included given the rarity of ICUCA. Recently a consensus statement from the American Hearts association suggested strategies for improving survival from inhospital cardiac arrest.6 The recommendations include strategies for ensuring rapid defibrillation with automated external defibrillators, prompt application of CPR and strategies for post resuscitation care and prognostication. Given the paucity of data on ICUCA it is unclear whether this strategy is applicable in the ICU setting.

4.1. Limitations The present review included studies that originated mostly from 1990s and 2000s. Most of the studies were conducted retrospectively in a single institution. Populations of the majority of the included studies were relatively small. Reviewed studies were designed and conducted in different areas of the world with different medical practices, policies and with considerable differences in between general population’s disease profile. Data collection and reporting were not standardized and reporting of outcome data remained scarce. It is important to mention, that 94% of the population included in our review originated from one large US multicenter study.9 In this population survival to hospital discharge was 15.9%. We decided not to perform a quantitative comparative data analysis, as we believe that due to the significant

I. Efendijev et al. / Resuscitation 85 (2014) 472–479

heterogeneity of the included studies it would not add any useful information to our review. 5. Conclusions At present data on intensive care unit cardiac arrest is quite limited and originates mostly from retrospective single center studies. The quality of data overall seems to be moderate at the most and focused prospective multi-center studies are needed. Data on incidence, outcomes and causes of death are highly variable. Conflict of interest statement None of the authors have any conflicts of interest, financial or otherwise, relevant to the content of this article. Acknowledgement Appreciation is expressed to Per Rosenberg professor emeritus of the University of Helsinki for his critical review of the manuscript. References 1. Myrianthefs P, Kalafati M, Lemonidou C, et al. Efficacy of CPR in a general, adult ICU. Resuscitation 2003;57:43–8. 2. Yi HJ, Kim YS, Ko Y, Oh SJ, Kim KM, Oh SH. Factors associated with survival and neurological outcome after cardiopulmonary resuscitation of neurosurgical intensive care unit patients. Neurosurgery 2006;59:836–8. 3. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009;339:6. 4. Cummins RO, Chamberlain D, Hazinski MF, et al. Recommended guidelines for reviewing, reporting, and conducting research on in-hospital resuscitation: the in-hospital “Utstein style”. A statement for healthcare professionals from the American Heart Association, the European Resuscitation Council, the Heart and Stroke Foundation of Canada, the Australian Resuscitation Council, and the Resuscitation Councils of Southern Africa. Resuscitation 1997;34:151–83. 5. Jennett B, Bond M. Assessment of outcome after severe brain damage. Lancet 1975;1:480–4. 6. Morrison LJ, Neumar RW, Zimmerman JL, et al. Strategies for improving survival after in-hospital cardiac arrest in the United States: 2013 consensus recommendations: a consensus statement from the American Heart Association. Circulation 2013;127:1538–63. 7. Enohumah KO, Moerer O, Kirmse C, Bahr J, Neumann P, Quintel M. Outcome of cardiopulmonary resuscitation in the intensive care units of a university hospital. Afr J Reprod Health 2006;10:161–70. 8. Enohumah KO, Moerer O, Kirmse C, Bahr J, Neumann P, Quintel M. Outcome of cardiopulmonary resuscitation in intensive care units in a university hospital. Resuscitation 2006;71:161–70. 9. Tian J, Kaufman DA, Zarich S, et al. American Heart Association National Registry for Cardiopulmonary Resuscitation I. Outcomes of critically ill patients who received cardiopulmonary resuscitation. Am J Respir Crit Care Med 2010;182:501–6. 10. Kutsogiannis DJ, Bagshaw SM, Laing B, Brindley PG. Predictors of survival after cardiac or respiratory arrest in critical care units. CMAJ 2011;183: 1589–95.

479

11. Chang SH, Huang CH, Shih CL, et al. Who survives cardiac arrest in the intensive care units? J Crit Care 2009;24:408–14. 12. Grigoriyan A, Vazquez R, Palvinskaya T, et al. Outcomes of cardiopulmonary resuscitation for patients on vasopressors or inotropes: a pilot study. J Crit Care 2009;24:415–8. 13. Lee HK, Lee H, No JM, et al. Factors influencing outcome in patients with cardiac arrest in the ICU. Acta Anaesthesiol Scand 2013;57:784–92. 14. Karetzky M, Zubair M, Parikh J. Cardiopulmonary resuscitation in intensive care unit and non-intensive care unit patients. Immediate and long-term survival. Arch Intern Med 1995;155:1277–80. 15. Landry FJ, Parker JM, Phillips YY. Outcome of cardiopulmonary resuscitation in the intensive care setting. Arch Intern Med 1992;152:2305–8. 16. Peterson MW, Geist LJ, Schwartz DA, Konicek S, Moseley PL. Outcome after cardiopulmonary resuscitation in a medical intensive care unit. Chest 1991;100:168–74. 17. Tortolani AJ, Risucci DA, Rosati RJ, Dixon R. In-hospital cardiopulmonary resuscitation: patient, arrest and resuscitation factors associated with survival. Resuscitation 1990;20:115–28. 18. Wallace S, Ewer MS, Price KJ, Feeley TW. Outcome and cost implications of cardiopulmonary resuscitation in the medical intensive care unit of a comprehensive cancer center. Support Care Cancer 2002;10:425–9. 19. Skrifvars MB, Varghese B, Parr MJ. Survival and outcome prediction using the Apache III and the out-of-hospital cardiac arrest (OHCA) score in patients treated in the intensive care unit (ICU) following out-of-hospital, in-hospital or ICU cardiac arrest. Resuscitation 2012;83:728–33. 20. Guney MR, Ketenci B, Yapici F, et al. Results of treatment methods in cardiac arrest following coronary artery bypass grafting. J Card Surg 2009;24: 227–33. 21. Smith DL, Kim K, Cairns BA, Fakhry SM, Meyer AA. Prospective analysis of outcome after cardiopulmonary resuscitation in critically ill surgical patients. J Am Coll Surg 1995;180:394–401. 22. Rabinstein AA, McClelland RL, Wijdicks EF, Manno EM, Atkinson JL. Cardiopulmonary resuscitation in critically ill neurologic-neurosurgical patients. Mayo Clin Proc 2004;79:1391–5. 23. Anthi A, Tzelepis GE, Alivizatos P, Michalis A, Palatianos GM, Geroulanos S. Unexpected cardiac arrest after cardiac surgery: incidence, predisposing causes, and outcome of open chest cardiopulmonary resuscitation. Chest 1998;113: 15–9. 24. Chan PS, Khalid A, Longmore LS, Berg RA, Kosiborod M, Spertus JA. Hospital-wide code rates and mortality before and after implementation of a rapid response team. JAMA 2008;300:2506–13. 25. Chan PS, Jain R, Nallmothu BK, Berg RA, Sasson C. Rapid response teams: a systematic review and meta-analysis. Arch Intern Med 2010;170:18–26. 26. Meaney PA, Nadkarni VM, Kern KB, Indik JH, Halperin HR, Berg RA. Rhythms and outcomes of adult in-hospital cardiac arrest. Crit Care Med 2010;38:101–8. 27. Sandroni C, Nolan J, Cavallaro F, Antonelli M. In-hospital cardiac arrest: incidence, prognosis and possible measures to improve survival. Intensive Care Med 2007;33:237–45. 28. Ringh M, Herlitz J, Hollenberg J, Rosenqvist M, Svensson L. Out of hospital cardiac arrest outside home in Sweden, change in characteristics, outcome and availability for public access defibrillation. Scand J Trauma Resus Emerg Med 2009;17:18. 29. Agarwal DA, Hess EP, Atkinson EJ, White RD. Ventricular fibrillation in Rochester, Minnesota: experience over 18 years. Resuscitation 2009;80:1253–8. 30. Rea TD, Pearce RM, Raghunathan TE, et al. Incidence of out-of-hospital cardiac arrest. Am J Cardiol 2004;93:1455–60. 31. Cobb LA, Fahrenbruch CE, Olsufka M, Copass MK. Changing incidence of out-of-hospital ventricular fibrillation, 1980–2000. JAMA 2002;288: 3008–13. 32. Anon. United Kingdom transient ischaemic attack (UK-TIA) aspirin trial: interim results. UK-TIA Study Group. Br Med J (Clin Res ed) 1988;296: 316–20.