Article Text

other Versions

Download PDFPDF

Age- and gender-specific differences in the prognostic value of CT coronary angiography
  1. Kai Hang Yiu1,2,
  2. Fleur R de Graaf1,
  3. Joanne D Schuijf1,
  4. Jacob M van Werkhoven1,2,
  5. Nina Ajmone Marsan1,
  6. Caroline E Veltman1,
  7. Albert de Roos3,
  8. Aju Pazhenkottil4,
  9. Lucia J Kroft3,
  10. Eric Boersma5,
  11. Bernhard Herzog4,
  12. Melissa Leung6,
  13. Erica Maffei7,
  14. Dominic Y Leung6,
  15. Philipp A Kaufmann4,8,
  16. Filippo Cademartiri7,
  17. Jeroen J Bax1,
  18. J Wouter Jukema1,2
  1. 1Department of Cardiology, Leiden University Medical Center, Leiden, The Netherlands
  2. 2The Interuniversity Cardiology Institute of the Netherlands, Utrecht, The Netherlands
  3. 3Department of Radiology, Leiden University Medical Center, Leiden, The Netherlands
  4. 4Department of Cardiology, University Hospital Zurich, Zurich, Switzerland
  5. 5Department of Cardiology, Erasmus Medical Center, Rotterdam, The Netherlands
  6. 6Department of Cardiology, Liverpool Hospital, University of New South Wales, Sydney, Australia
  7. 7Department of Radiology and Cardiology, Azienda Ospedaliero-Universitaria, Parma, Italy
  8. 8Zurich Integrative Human Physiology, University of Zurich, Zurich, Switzerland
  1. Correspondence to Dr J Wouter Jukema, Department of Cardiology, Leiden University Medical Center, Albinusdreef 2, 2333 ZA Leiden, The Netherlands; j.w.jukema{at}lumc.nl

Abstract

Objective To evaluate the potential age- and gender-specific differences in the incidence and prognostic value of coronary artery disease (CAD) in patients undergoing CT coronary angiography (CTA).

Design and patients In this multicentre prospective registry study, 2432 patients (mean age 57±12, 56% male) underwent CTA for suspected CAD. Patients were stratified into four groups according to age <60 or ≥60 years and, male or female gender.

Main outcome measures A composite end point of cardiac death and non-fatal myocardial infarction.

Results CTA results were normal in 991 (41%) patients, showed non-significant CAD in 761 (31%) patients and significant CAD in the remaining 680 (28%) patients. During follow-up (median 819 days, 25–75th centile 482–1142) a cardiovascular event occurred in 59 (2.4%) patients. The annualised event rate was 1.1% in the total population (men=1.3% and women=0.9%). In patients aged <60 years, the annualised event rate of male and female patients was 0.6% and 0.5%, respectively. Among patients aged ≥60 years the annualised event rate was 1.9% in male and 1.1% in female patients. Observations on CTA predicted events in male patients, both age <60 and ≥60 years and in female patients age ≥60 years (log-rank test in all groups, p<0.01). However, CTA provided limited prognostic value in female patients aged <60 years (log-rank test, p=0.45).

Conclusions After age and gender stratification, CTA findings were shown to be of limited predictive value in female patients aged <60 years as compared with male patients at any age and female patients aged ≥60 years.

  • Atherosclerosis
  • CT coronary angiography
  • gender
  • atherosclerosis
  • transthoracic
  • aortic valve disease
  • cardiac imaging
  • CT scanning
  • nuclear cardiology
  • lipoproteins
  • lipid drugs
  • drug trials
  • lipids in atherosclerosis

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Coronary artery disease (CAD) is the leading cause of mortality both in men and women. However, the incidence of obstructive CAD differs between genders; in premenopausal women, the incidence of obstructive CAD appears to be approximately one-third that of age-matched men.1 With increasing age this difference tends to diminish with similar incidences of obstructive CAD in men and women aged ≥75 years.1 Similar findings have been reported for the extent of coronary plaque burden using both invasive and non-invasive imaging modalities.2 3

CT coronary angiography (CTA) is an emerging non-invasive tool for detecting obstructive CAD with high diagnostic accuracy.4 In addition to the detection of obstructive coronary artery stenosis, however, the techniques allows non-invasive visualisation of non-obstructive CAD and plaque composition and significant coronary artery stenosis.5 6 Moreover, studies have shown that CTA can provide important prognostic information and risk stratification in patients with suspected CAD.7–9 However, these studies have mainly focused on the general population and limited data are available for gender-specific differences.

Recently, Shaw et al showed, using 16-slice CTA, that evaluation of CAD on CTA added incremental value to clinical assessment for risk stratification both in men and women.3 Interestingly, the extent of non-obstructive CAD detected by CTA predicted mortality in women but not in men. However, the effect of age was not taken into account in this particular study. Given the fact that differences in coronary atherosclerosis between genders are highly age-dependent, it is conceivable that age may substantially influence the predictive value of CTA in women as compared with men. Therefore the aim of this study was to evaluate the potential age- and gender-specific differences of CAD incidence and the prognostic value in patients undergoing CTA. Accordingly, the frequency of CAD and the predictive value for cardiovascular (CVS) events by CTA was investigated in four subpopulations stratified according to gender (male or female) and age (aged <60 or ≥60 years).

Methods

Study population

The study population consisted of 2474 patients who underwent CTA for suspected CAD. Patients were consecutively enrolled as part of an ongoing registry assessing the predictive value of CTA for CVS events retrospectively at (1) the University of Zurich, Switzerland; (2) the Leiden University Medical Center, The Netherlands; (3) Azienda Ospedaliero-Universitaria, Parma, Italy and (4) Liverpool Hospital, University of New South Wales, Liverpool, New South Wales, Australia. Exclusion criteria included cardiac arrhythmias, renal insufficiency (defined as a glomerular filtration rate <30 ml/min/1.73 m2), known hypersensitivity to iodine contrast media, pregnancy, previous percutaneous coronary intervention and myocardial infarction. Moreover, a total of 42 uninterpretable scans were excluded from the analysis (33 owing to motion artefact/poor scan quality, eight owing to high coronary calcification leading to blooming artefact and one owing to non-diagnostic CTA stent image quality). Accordingly, 2432 patients were included in the final analysis. The pre-test likelihood of obstructive CAD was evaluated using Diamond and Forrester's criteria.10 Patients were entered prospectively into the departmental patient information system and retrospectively analysed. This study was approved by the local ethics committees in all participating centres and all patients provided informed consent.

CT coronary angiography protocol

Examinations were performed using one of the following scanners: (1) 320-row CTA scanner (Aquilion ONE, Toshiba Medical Systems, Otawara, Japan) in 97 (4%) patients; (2) 64-row CTA scanner (Lightspeed VR 64, General Electrics (GE), Milwaukee, Michigan, USA or Aquillion64, Toshiba Medical Systems, Tokyo Japan or Sensation64, Siemens, Forchheim, Germany) in 2335 (96%) patients. Patients' heart rate and blood pressure were monitored before each CTA scan. In the absence of contraindications, 44% patients with a heart rate >65 beats/min were given β-blocking drugs (50–100 mg metoprolol, orally or 5–10 mg metoprolol, intravenously) for heart rate optimisation. A description of all scan parameters has been published previously.6 11–13 In brief, all images were acquired during a single inspiratory breath-hold of maximally 12 s. For 320-row CTA, the ECG was registered simultaneously for prospective triggering of the data.6 The entire heart was imaged in a single heart beat and maximal tube current was attained during 75% of the R-R interval in patients with stable heart rate <60 beats per minute (bpm), during 65–85% of the R-R interval in patients with a heart rate 60–65 bpm or during 30–80% of the R-R interval in patients with a heart rate >65 bpm. In addition, a collimation of 320×0.5 mm was used. Additional scan parameters were as follows: 350 ms gantry rotation time, 120–135 kV tube voltage and 400–580 mA (depending on body mass index and thoracic geometry). For 64-row CTA, a helical-scanning technique was used as previously described, using a collimation of 64×0.5 mm.11–14 Specifically, retrospective gating of the data was registered by simultaneous ECG monitoring. When prospective triggering was performed, a small acquisition window during mid-diastolic phase of the R-R cycle was used (ie, 70–80% in the case of low heart rate).14 Retrospective gating of the data was registered by the simultaneous ECG monitoring, and a collimation of 64×0.5 mm was used. Additional scan parameters were 400 ms or 500 ms gantry rotation time depending on the cardiac frequency, 120 kV tube voltage and 300–350 mA (depending on body mass index and thoracic geometry).

The estimated mean radiation dose for retrospective gating with 64-slice CT is 18.0 mSv (range 6.8–45.5 mSv).15 For prospective triggering with 64-slice CTA the estimated mean radiation dose is 2.1 mSv (range 1.1–3.0 mSv).14 For prospective triggering using 320-row CTA, the estimated mean radiation dose is 3.9 mSv (range 2.7–26.2 mSv).6

Data analysis

Post-processing of the CTA was performed on dedicated workstations (Vitrea2, Vital Images, Minneapolis, Minnesota, USA or Advantage, GE Healthcare, Waukesha, Wisconsin, USA or Syngo InSpace4D Application, Siemens, Munich, Germany or Aquarius, TeraRecon, San Mateo, California, USA). Coronary anatomy was assessed using a standardised method by dividing the coronary arteries into 17 segments according to the modified American Heart Association classification.16 CTA results were defined as normal CTA (no identifiable plaque or minimal wall irregularities), non-obstructive CAD (<50% luminal narrowing), or obstructive CAD (≥50% luminal narrowing).

Follow-up of patients

Follow-up data were collected by clinical visits or standardised telephone interviews with each patient or his or her direct relative, or with the referring doctor to discuss symptoms, the occurrence of new events or change in clinical status, and any hospital admission. A composite of CVS end point was defined as non-fatal myocardial infarction, and cardiac death (congestive heart failure, fatal myocardial infarction or sudden cardiac death). Non-fatal myocardial infarction was defined based on criteria of typical angina, elevated cardiac enzyme levels and typical changes on the ECG.17

Statistical analysis

Continuous variables are presented as mean±SD and compared using either the Student t test or Wilcoxon's rank-sum test, as appropriate. Categorical data are presented as frequencies and percentages and compared using the χ2 or the Fisher exact test if at least one cell had an expected cell count <5. Kaplan–Meier curves were constructed and the outcomes of CTA results in different patient subgroups were compared using the log-rank test. All statistical analyses were performed using the statistical package SPSS for windows (V.15.0, SPSS). A p value <0.05 was considered to be statistically significant.

Results

Baseline clinical characteristics

The study population consisted of 2432 patients presenting with suspected CAD at the University of Zurich (n=391), Leiden University Medical Center (n=758), Azienda Ospedaliero-Universitaria di Parma (n=1169) and Liverpool Hospital (n=114). The baseline characteristics of the patient population are shown in table 1. The average age of all patients was 56.5±12.4 years and 56% of them were male. Comparing between genders, male patients were more likely to have a higher incidence of diabetes and smoking history while female patients were likely to be older, have a higher incidence of hypertension and family history of CAD. A total of 2097 (86%) were referred for CTA owing to the presence of symptoms while the remaining 335 (14%) patients were asymptomatic. For both male and female patients, the majority presented with an intermediate pre-test likelihood while female patients had a slightly higher frequency of low pre-test likelihood.

Table 1

Clinical characteristics of study population

CT coronary angiography results

For the whole study population, 991 (41%) patients had a normal CTA, 761 (31%) had non-obstructive CAD and 680 (28%) had obstructive CAD. Male patients had a higher incidence of obstructive CAD (34% vs 21%, p<0.01) and female patients had a higher incidence of normal CTA results (49% vs 34%, p<0.01) as shown in figure 1. The CTA results of the four subgroups stratified according to age and gender are shown in figure 2. Male patients aged ≥60 years had more extensive CAD with nearly half of patients having obstructive CAD and less than 20% of patients having a normal CTA. On the other hand, two-thirds of female patients aged <60 years had a normal CTA result while only 11% had obstructive CAD.

Figure 1

Results of CT coronary angiography (CTA) in (A) male and (B) female patients. CAD, coronary artery disease.

Figure 2

Results of CT coronary angiography (CTA) for the detection of non-obstructive and obstructive coronary artery disease (CAD) in four subgroups of patients: (A) man aged <60 years; (B) man aged ≥60 years; (C) woman aged <60 years; (D) woman aged ≥60 years.

Follow-up results

The median follow-up was 819±335 (25th–75th centile 482–1142) days. A CVS event occurred in 59 patients (2.4%) including non-fatal myocardial infarction in 34 cases and CVS mortality in 25 cases. No difference in the occurrence of the composite end point was noted between male (aged <60=10 events, aged ≥60=28 events) and female (aged <60=5 events, aged ≥60=16 events) patients (38 events (2.8%) vs 21 events (2.0%), p=0.23).

Event rates

The annualised event rate was 1.1% in the total population (men=1.3% and women=0.9%). In patients aged <60 years, the annualised event rate of male and female patients was 0.6% and 0.5%, respectively. Among patients aged ≥60 years an annualised event rate of 1.9% was observed in male and 1.1% in female patients.

Predictive value of CT coronary angiography results in different age and gender subgroups

The Kaplan–Meier survival curves comparing CTA results in men and women are shown in figure 3. In male patients, annualised event rates were 0.07%, 0.8% and 2.9% in the presence of a normal CTA, non-obstructive CAD and obstructive CAD, respectively. For women, these values were 0.2%, 1.2% and 2.1%, respectively. CTA findings were predictive of the composite end point for both male and female patients (log-rank test p<0.01 for both analyses). In addition, the presence of non-obstructive or obstructive CAD was predictive of the composite end point in both male (log-rank test p=0.03 and p<0.01, respectively) and female (log-rank test p<0.01 for both analyses) patients compared with normal CTA. Both male and female patients were further divided into two age groups (aged <60 years or ≥60 years) and the corresponding Kaplan–Meier survival curves are shown in figure 4.

Figure 3

Composite endpoint-free survival in patients with different gender according to CT coronary angiography (CTA) results. (A) Kaplan–Meier curves for composite end point in male patients with normal CTA, non-obstructive coronary artery disease (CAD) and obstructive CAD. (B) Kaplan–Meier curves for composite end point in female patients with normal CTA, non-obstructive CAD and obstructive CAD.

Figure 4

Composite endpoint-free survival in patients with different age and gender subgroups according to CT coronary angiography (CTA) results. (A) Kaplan–Meier curves for composite end point in male patients aged <60 years with normal CTA, non-obstructive coronary artery disease (CAD) and obstructive CAD. (B) Kaplan–Meier curves for composite end point in male patients aged ≥60 years with normal CTA, non-obstructive CAD and obstructive CAD. (C) Kaplan–Meier curves for composite end point in female patients aged <60 years with normal CTA, non-obstructive CAD and obstructive CAD. (D) Kaplan–Meier curves for composite end point in female patients aged ≥60 years with normal CTA, non-obstructive CAD and obstructive CAD.

In male patients aged <60 years, the annualised event rates in the presence of normal CTA, non-obstructive CAD and obstructive CAD were 0.1%, 0.4% and 6.3%, respectively. Among male patients aged ≥60 years, annualised event rates increased from 0% for normal coronary arteries to 1.1% in the presence of non-obstructive CAD to 3.2% in the presence of obstructive CAD. In male patients aged <60 years or ≥60 years, the predictive role of CTA was significant (log-rank test p<0.01, p<0.01 for both groups). When compared with normal CTA, the presence of obstructive CAD was significantly predictive of the composite end point at both age groups (aged <60 years or ≥60 years) (log-rank test p<0.01, p<0.01 respectively). In contrast, non-obstructive CAD was not predictive of CVS events (log-rank test p=0.35 and p=0.11, respectively).

In female patients aged <60 years, the annualised event rates in the presence of normal CTA, non-obstructive CAD and obstructive CAD were 0.3%, 0.9% and 0.7%, respectively. Among female patients aged ≥60 years, annualised event rates increased from 0% for normal coronary arteries to 1.4% in the presence of non-obstructive CAD to 2.4% in the presence of obstructive CAD. In female patients aged ≥60 years, CTA results were predictive of CVS events (log-rank test p<0.01). Moreover, both non-obstructive and obstructive CAD were predictive of the composite end point when compared with normal CTA (log-rank test p<0.01 and p<0.01, respectively). Interestingly, in female patients aged <60 years, the role of CTA in predicting CVS events was not significant (log-rank test p=0.45). Further analysis showed that neither the presence of non-obstructive nor obstructive CAD was predictive of CVS events in comparison with normal CTA (log-rank test p=0.25 and p=0.35, respectively).

Discussion

This study is one of the first to evaluate gender-specific differences in the incidence of CAD on CTA as well as in the predictive role of CTA for CVS events with respect to age. After age stratification (<60 years or ≥60 years), observations on CTA remained predictive for CVS events in male patients regardless of age. Similarly, CTA was a strong predictor of CVS events in female patients aged ≥60 years. In contrast, however, CTA provided limited prognostic value in female patients aged <60 years.

Prevalence of CAD between genders in relation to age

In concordance with previous studies, this investigation showed that in patients referred for CTA, male patients had a substantially higher prevalence of obstructive CAD than female patients.3 18 19 Interestingly, male patients in our cohort were likely to have a higher prevalence of diabetes mellitus and smoking history, but were younger and with a lower prevalence of hypertension as compared to female patients. Indeed, a recent study by Faletra and colleagues showed that male gender, older age, diabetes and hypercholesterolaemia independently predicted the presence of coronary artery plaque in 920 patients without a previous history of CAD.20 Therefore, the clustering of several CVS risk factors in male patients may further contribute to a higher prevalence of obstructive CAD as observed in our cohort. Female patients aged <60 years had a lower overall prevalence of both non-obstructive and obstructive CAD than male patients. In elderly patients (aged ≥60 years), however, the rate of non-obstructive CAD became similar between genders, although the rate of obstructive CAD remained higher in male patients as compared with female patients. These observations are in line with previous literature examining age and gender differences in CAD.21–24 Indeed, several pathological and imaging studies have confirmed that in women an initial lag of approximately 10 years in the development of CAD is present.2 21 Potentially, this delay in CAD development might be attributed to the protective effect of premenopausal female hormone as the difference in CAD prevalence between genders tends to disappear at the start of the sixth decade.25

Prognostic value of CTA between genders

In addition to providing an accurate diagnosis for CAD, recent studies have demonstrated that CTA has independent prognostic value for CVS events in patients with suspected CAD.7–9 A recent meta-analysis with 9592 patients who underwent CTA demonstrated that future CVS events increase with increasing severity of CAD. In accordance with this, our study also demonstrated the prognostic value of CTA in male and female patients.26

Shaw et al demonstrated that the presence of obstructive CAD had significant predictive value for future CVS events in both genders.3 In addition, the presence of non-obstructive CAD was only predictive for future CVS events in female patients but not in male patients. In concordance with their findings, this investigation demonstrated that the presence of obstructive CAD had significant prognostic value for both genders. In contrast to their study, our study showed that non-obstructive CAD was predictive for future CVS events in female patients and also in male patients. Conceivably, non-obstructive plaque may play a more important role in female patients, whereas its predictive value may be less in male patients. Accordingly, the discrepancy between the study by Shaw et al and ours may be explained by the larger number of patients enrolled in our study. As compared with female patients, male patients had a larger separation of survival curves between non-obstructive and obstructive CAD in comparison with normal CTA. This observation could be partially explained by the altered coronary reactivity that frequently occurred in women.27 In addition, a recent study by Pepine et al showed that coronary microvascular reactivity to adenosine provided prognostic value over angiographic CAD severity in female patients.28 Nevertheless, the underlying mechanism of non-obstructive CAD leading to more adverse outcome in female patients than in male patients with non-obstructive CAD will require further study.

In concordance with data from the National Institutes of Health, National Heart, Lung and Blood Institute, a low annualised CVS event rate was seen in female patients aged <60 years in comparison with male patients with similar age.29 The low incidence of CVS events seen in young female patients aged <60 years may partly be attributed by the protective female sex hormone profile in the premenopausal state.30 31 Indeed, there is a paucity of data evaluating the age-specific prognostic role of CTA in male and female patients. Accordingly, patients in this cohort were stratified according to age (<60 years or ≥60 years old) to evaluate the impact of age on the prognostic role of CTA between genders. For male patients aged <60 years or ≥60 years, the presence of obstructive CAD was significantly predictive for future CVS events. In female patients aged ≥60 years, the presence of non-obstructive CAD was shown to have a significant prognostic role in addition to obstructive CAD. Conversely, the prognostic value for CVS events by CTA was limited in female patients aged <60 years. Possibly, other CVS factors, such as diabetes may have greater prognostic impact in female patients than the presence of CAD alone.32 Accordingly, while CTA may be a valuable technique to rule out CAD and avoid unnecessary invasive coronary angiography, its prognostic value appears to be limited in women aged <60 years.

Limitations

The results of CTA were only evaluated visually, although there is a lack of accurate and consensual quantitative algorithms. Patients with established CAD including previous myocardial infarction or percutaneous coronary intervention were excluded. Moreover, only patient-basis data was recorded and therefore analysis of individual segments and location of disease was not available in this study. The overall events rate was low in this study, which may be related to the use of only hard clinical end points (non-fatal myocardial infarction and CVS deaths) and a significant number of patients with low pre-test likelihood. Despite the incremental prognostic value of CTA, in particular the high negative predictive value of normal CTA, a recent study has shown that the CVS events rate remains low even in patients with positive CTA results.26 Nevertheless, the use of CTA may identify those with negative CTA results who require no further testing and those with extensive disease who require immediate referral and treatment. The ongoing SPARC trial, a prospective, multicentre, observational registry, aims to evaluate the prognostic value and post-test resource use of multiple imaging modalities, including CTA.33 On the other hand, the current guideline for the use of CTA is based on expert consensus, and thus large-scale randomised studies are needed to evaluate the clinical effectiveness guided by the test results.34 Multivariate adjustment for the independent prognostic role of CTA results was not performed owing to a lack of statistical power as a result of the limited cardiovascular events in each subgroup after age and gender stratification. Finally, it is important to realise that despite the reduced radiation burden due to prospective ECG triggering, the risk of radiation exposure associated with CTA should not be disregarded, particularly in younger women.

Conclusion

This study confirmed that male patients had a significantly higher frequency of obstructive CAD on CTA than female patients. Moreover, the presence of both non-obstructive CAD and obstructive CAD on CTA was significantly predictive of CVS events in both genders. Importantly, after age stratification, CTA findings were shown to be of limited predictive value in female patients aged <60 years as compared with male patients at any age and female patients aged ≥60 years.

References

Footnotes

  • Funding KHY receives reseach grants from the Hong Kong Heart Foundation. JWJ receives research grants from and was a speaker at meetings sponsored by Astellas, AstraZeneca, Biotronic, Boston Scientific, Bristol-Myers Squibb, Cordis, Daiichi Sankyo, Eli Lilly and Company, Medtronic, Merck-Schering Plough, Pfizer, Orbus Neich, Novartis, Roche, Servier, the Netherlands Heart Foundation, the Interuniversity Cardiology Institute of the Netherlands and the European Community framework KP7 program. JJB receives grants from Biotronik, BMS medical imaging, Boston Scientific, Edwards Lifesciences, GE Healthcare, Medtronic and St Jude Medical. JvW is financially supported by a research grant from the Netherlands Society of Cardiology (Utrecht, The Netherlands). PAK is supported by a grant from the Swiss National Science Foundation (Berne, Switzerland) (SNSF-professorship grant nr.PPOOA-114706), The remaining authors: None.

  • Competing interests None.

  • Ethics approval This study was approved by the local ethics committees in all participating centres and all patients provided informed consent.

  • Provenance and peer review Not commissioned; externally peer reviewed.