CRD summary

This review found that the Global Registry in Acute Coronary Events (GRACE) score had better performance than the Thrombolysis in Myocardial Infarction (TIMI) score when predicting acute coronary syndrome events. This review included a very large number of patients so, despite some concerns over the diversity of studies, the authors’ conclusions are probably reliable.

Authors' objectives

To assess the performance of risk scores for the prediction of acute coronary syndromes (ACS).

Searching

MEDLINE was searched for studies published in English. Search terms were presented. Meeting abstracts (not specified) and references from included studies were searched.

Study selection

Eligible studies were derivation and validation studies of risk scores for ACS. ACS consisted of unstable angina and ST and non-ST segment elevation myocardial infarction (STEMI and NSTEMI). Studies had to investigate patients who presented at hospital with ACS, appraise risk scores using multivariate analysis and present a prediction score.

The most commonly assessed risk scores were Thrombolysis in Myocardial Infarction (TIMI) and Global Registry in Acute Coronary Events (GRACE) scores. About two-thirds of all patients were male and nearly all were aged in their 60s. There was considerable variation in numbers of patients who presented with ACS: median proportions with unstable angina ranged from 21% to 54% and with NSTEMI ranged from 31% to 78%.

Two reviewers independently performed the study selection. Disagreements were resolved by consensus.

Assessment of study quality

Study quality was assessed using a modified version of the MOOSE (meta-analysis of observational studies in epidemiology) guidelines. Aspects considered included study design and setting, data source, statistical methods and risk of analytical, selection, adjudication detection and attrition biases.

Two reviewers independently performed the quality assessment. Disagreements were resolved by consensus.

Data extraction

The area under the curve (AUC) or the concordance index (c-index) were extracted for each risk score in each study, along with its corresponding 95% confidence interval (CI).

Two reviewers independently performed data extraction. Disagreements were resolved by consensus.

Methods of synthesis

Pooled AUC with 95% CI were calculated using random-effects inverse-variance meta-analysis. Results were compared to a fixed-effect model. Small-study bias was investigated using funnel plots.

Results of the review

Eighty-two studies were included in the review. Follow-up times appeared to range from 30 to 392 days.The percentages of studies judged to be at low risk of bias were approximately 65% for selection bias, 45% for attrition bias, 41% for adjudication bias and 54% for analytical bias. Most of the rest were judged to be at moderate or unclear risk of bias.

ACS: There were seven derivation studies (25,525 patients). In these studies short-term TIMI had an AUC of 0.66 (95% CI 0.64 to 0.68) and short-term GRACE had an AUC of 0.83 (95% CI 0.82 to 0.84). There were 15 validation studies (257,654 patients). In these studies short-term TIMI had an AUC of 0.73 (95% CI 0.69 to 0.78) and short-term GRACE had an AUC of 0.82 (95% CI 0.80 to 0.89).

Unstable angina and NSTEMI: There were 18 derivation studies (56,560 patients) and 18 validation studies (56,673 patients). In the validation studies short-term TIMI had an AUC of 0.54 (95% CI 0.52 to 0.57) and short-term GRACE had an AUC of 0.83 (95% CI 0.79 to 0.87). Long-term TIMI had an AUC of 0.67 (95% CI 0.62 to 0.71) and long-term GRACE had an AUC of 0.80 (95% CI 0.74 to 0.89).

STEMI studies: There were 15 derivation studies (134,557 patients) and 17 validation studies (187,619 patients). Short-term TIMI had an AUC of 0.77 (95% CI 0.71 to 0.83) and short-term GRACE an AUC of 0.82 (95% CI 0.81 to 0.83). Long-term TIMI had an AUC of 0.77 (95% CI 0.72 to 0.85) and long-term GRACE had an AUC of 0.81 (95% CI 0.80 to 0.82).

Results for other risk scores were presented but data on these was very limited. There was no evidence of small-study bias.

Authors' conclusions

GRACE performed better than TIMI when predicting ACS. Other risk scores were available but had not undergone rigorous validation.

CRD commentary

This was generally a well-conducted review with appropriate inclusion criteria. The search was limited to only one database and studies published in English so some relevant studies may have been missed but a substantial number of cohorts and patients were included. Action was taken to reduce reviewer error and bias throughout the review process. Study quality was assessed and the risk of bias was judged to be generally low to moderate. Most of the included studies were observational cohorts and so may be subject to some bias.

Meta-analyses were conducted to synthesise the studies. The authors noted considerable clinical diversity across the studies, particularly between derivation and validation studies, in outcomes predicted and in follow-up times. Heterogeneity assessments were not presented in detail in the review so it was uncertain how this diversity may have affected the results and how generalisable these results might be to different situations. Despite this uncertainty the results appear to be broadly consistent and, given the large size of this review, the authors' conclusions are probably reliable.

Implications of the review for practice and research

Practice: The authors made no recommendations for medical practice.

Research: The authors suggested that research was needed to assess the performance of scores other than GRACE and TIMI.

Funding

Not stated.

Bibliographic details

D'Ascenzo F, Biondi-Zoccai G, Moretti C, Bollati M, Omede P, Sciuto F, Presutti DG, Modena MG, Gasparini M, Reed MJ, Sheiban I, Gaita F. TIMI, GRACE and alternative risk scores in acute coronary syndromes: a meta-analysis of 40 derivation studies on 216,552 patients and of 42 validation studies on 31,625 patients. Contemporary Clinical Trials 2012; 33(3): 507-514. [PubMed: 22265976]

Original Paper URL

http://www.sciencedirect.com/science/article/pii/S1551714412000195

Indexing Status

Subject indexing assigned by NLM

MeSH

Acute Coronary Syndrome /diagnosis /pathology /therapy; Aged; Angioplasty, Balloon, Coronary; Area Under Curve; Confidence Intervals; Decision Making; Female; Humans; Male; Middle Aged; Prognosis; Risk Assessment /methods

AccessionNumber

12012018188

Database entry date

17/05/2013

Record Status

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.