Measuring outcome after cardiac arrest: construct validity of Cerebral Performance Category

Abstract

Introduction

Approximately half of the survivors of cardiac arrest have cognitive impairments due to hypoxic brain injury. To describe the outcome after a cardiac arrest, the Cerebral Performance Category (CPC) is frequently used. Although widely used, its validity is still debatable.

Objective

To investigate the construct validity of the Cerebral Performance Category in survivors of a cardiac arrest.

Participants were 18 years and older that survived a cardiac arrest more than six months.

Methods

Cross-sectional design. A method to administer the CPC in a structured and reproducible manner was developed. This ‘Structured CPC’ was administered by a structured interview. Construct variables were Cognitive Failure Questionnaire (CFQ), Barthel Index (BI), Frenchay Activity Index (FAI), Community Integration Questionnaire (CIQ) and Quality of Life after Brain Injury (Qolibri). Associations were tested based on Spearman correlation coefficients.

Results

A total of 62 participants responded. In 58 (94%) patients the CPC was determined, resulting in CPC 1 (48%), CPC 2 (23%) and CPC 3 (23%). The CPC-scoring correlated significantly with the CFQ (r = −0.40); BI (r = −0.57); FAI (r = −0.65), CIQ (r = −0.53) and Qolibri (r = −0.67).

Discussion and conclusions

In this study we developed the ‘Structured CPC’ to improve the transparency and reproducibility of the original CPC. A moderate correlation between the ‘Structured CPC’ and the constructs ‘activities’, ‘participation’ and ‘quality of life’ confirmed the validity of the ‘Structured CPC’.

Clinical message

The ‘Structured CPC’ can be used as an instrument to measure the level of functioning after cardiac arrest.

Introduction

The survival rate of out-of-hospital cardiac arrest to hospital discharge ranges between 6.7% and 8.4%.¹

During cardiac arrest the brain suffers from temporary limitation of the blood supply. This may lead to irreversible brain damage, called hypoxic–ischemic brain injury. Cognitive impairments are found in approximately half of the survivors.² Common symptoms of hypoxic–ischemic brain injury are disturbances of memory, attention and executive functioning.³

It is important to pay attention not only to survival but also to functional outcome after cardiac arrest. A frequently used instrument to determine outcome after cardiac arrest is the Cerebral Performance Category (CPC). The CPC consists of a 5-point scale (Table 1), in which CPC scores of 1 and 2 are mostly considered as ‘good’ outcomes and a CPC 3, 4 and 5 ‘poor’ outcomes.⁴ According to the Utstein criteria, it is recommended to collect neurological outcomes at discharge, by using the CPC or modified Ranking Scale (mRS).⁶ In scientific research, the CPC has been used as an outcome variable in many studies.7, 8, 9, 10 However, currently there is no standardized method to determine the CPC score. As a result, the procedure to score the CPC varies widely, for example by interpretation of retrospective chart reviews or asking persons by phone calls.

In addition, it is currently unclear what construct of functioning the CPC actually aims to assess.

The International Classification of Functioning, Disability and Health (ICF) is a classification of health and health-related domains.¹¹ The ICF distinguishes three different domains of daily life functioning: ‘body functions and structures’ (e.g. blood circulation and heart), ‘activities’ (e.g. standing) and ‘participation’ (e.g. working). It is unclear which of the three domains of the ICF model represents the CPC best.

Several researchers have tried to determine the construct of the CPC by testing its association with other instruments. Correlations between the CPC, determined by retrospective chart review, and the Functional Status Questionnaire, measuring all 3 domains of the ICF, were all low at discharge and moderate at follow-up.¹² A different study, also using a retrospective chart review, found a moderate correlation between CPC and modified Ranking Scale (mRS), measuring the domain ‘activities’.¹³ In addition, compared to the Health Utilities Index (HUI), measuring quality of life, a CPC 1 was related with a good HUI and a CPC > 1 ruled out a good quality of life.⁵ In this study the CPC was evaluated by an interview.

It is remarkable that in all studies mentioned a variety of instruments were used as the construct variable, but none of the studies identified a good relationship between the CPC and the construct variables chosen. Whether this finding indicates that CPC is indeed insufficient as a measure for functioning or quality of life or whether this finding is especially due to the variety of methods used to score the CPC is currently unclear.

This shows that the construct validity of the CPC is currently still debatable. In addition, the method to administer the CPC remains unclear.

The aim of this study was to investigate the construct validity of the CPC. Therefore, we first developed a semi-structured interview to assess the CPC in a systematic way in survivors of a cardiac arrest. The construct validity of this ‘Structured CPC’ was determined by comparing outcomes on the CPC with the constructs ‘body functions and structures’, ‘activities’, ‘participation’ of the ICF model and in addition the variable ‘quality of life’.

Based on the findings in earlier studies12, 13, a low correlation between CPC and the construct ‘body functions and structures’, a moderate correlation between the CPC and the ICF-constructs ‘activities’ and ‘participation’ and a low correlation between CPC and ‘quality of life’ were expected.