Transradial (TR) access for percutaneous coronary intervention (PCI) improves outcomes and reduces the risk of major bleeding compared with transfemoral (TF) access. However, data on gender-stratified outcomes based on vascular access are limited. Databases were queried to find relevant articles. Primary outcomes, including major bleeding complications, mortality, and secondary outcome including major adverse cardiovascular events (MACEs), myocardial infarction, and cerebrovascular accidents, were analyzed using a random-effect model to calculate unadjusted odds ratio (OR) of TR-PCI and TF-PCI between the genders. A total of 9 studies comprising 3,889,257 patients (389,580 in the TR arm and 3,499,677 in the TF arm) were included. Males comprised 73% and 67% of the TR and TF arms, respectively. TR-PCI was associated with lower major bleeding (pooled OR 0.51, 95% CI 0.40 to 0.64, p = 0.00; female OR 0.49, 95% CI 0.34 to 0.71, p = 0.00; male OR 0.54, 95% CI 0.40 to 0.73, p = 0.00) and mortality (pooled OR 0.54, 95% CI 0.45 to 0.66, p = 0.00; female OR 0.56, 95% CI 0.44 to 0.71, p = 0.27; male OR 0.54, 95% CI 0.39 to 0.75, p = 0.00) regardless of gender as compared with TF-PCI. Furthermore, TR-PCI also showed lower MACE (pooled OR 0.74, 95% CI 0.66 to 0.84, p = 0.00; female OR 0.64, 95% CI 0.59 to 0.70, p = 0.00; male OR 0.81, 95% CI 0.66 to 0.98, p = 0.00) as compared with TF-PCI in both genders. On analysis of interaction magnitude of the difference of favor of female and male for TR-PCI showed no statistically significant measurable difference. Periprocedural myocardial infarction and cerebrovascular accidents were not statistically different in TR and TF-PCI and were not different based on gender. In conclusion, TR-PCI was associated with a lower risk of major bleeding, mortality, and MACE irrespective of gender. In conclusion, TR-PCI should be the default access.
Transradial (TR) and transfemoral (TF) are common vascular routes of percutaneous coronary intervention (PCI). Major bleeding and vascular access complications remain the most common early complications of post-PCI and are associated with increased risk of mortality and major adverse cardiovascular events (MACEs). TR-PCI reduces the risk of major bleeding and vascular complications compared with TF-PCI, thereby reducing post-PCI mortality but interestingly, women are still twice as likely to experience major bleeding complications compared with men. Observational studies have demonstrated that post-PCI bleeding is more common in women even after controlling for bleeding risk factors driving for higher mortality when compared with men with bleeding. , Therefore, the TR-PCI is preferred as a bleeding avoidance strategy. , However, women were under-represented in landmark trials that compared TR and TF access. Furthermore, the gender-based outcomes were not thoroughly studied in major clinical trials. Therefore, the purpose of this study was to perform a systematic review and meta-analysis to investigate the gender-based cardiovascular outcomes of TR-PCI and TF-PCI.
A systematic search following the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses and guidelines of the Meta-Analysis of Observational Studies in Epidemiology , was performed ( Figure 1 ). Online bibliographic databases were used. Using boolean logic, a combination of MeSH terms “percutaneous coronary angiography,” “Transradial,” “Transfemoral,” “Stable angina,” “Unstable angina,” “STEMI” NSTEMI” were used. Previous publications published on the topic of this topic were included in a cross-reference check ( Supplementary Section 1 ).
Studies that included adult patients (age ≥18 years) who underwent PCI with either TR or TF access for angina (stable or unstable), ST or non–ST-segment elevation myocardial infarction (MI), and compared clinical outcomes based on gender were included. Studies with insufficient data, systematic reviews, meta-analyses, letters, editorials, case reports, conference abstracts, and case series with <10 patients were also excluded.
Two investigators (Y.S. and J.A.) removed duplicate studies using Endnote × 9 and independently screened titles and abstracts of all articles in the initial search. The senior investigator arbitrated discrepancies regarding the inclusion of studies. Studies were also screened by backward snowballing, wherein the references of included studies were used to guide searches. The Cochrane Risk of Bias tool and Newcastle-Ottawa Scale were used for quality assessment of included randomized controlled trials (RCTs) and observational studies. These risks of bias tools were evaluated by 2 investigators (Y.S. and J.A.) ( Supplementary Section 2 ).
The primary outcomes of interest were major bleeding complications and all-cause mortality. Major bleeding was defined as significant blood loss, including gastrointestinal bleeding, intracranial hemorrhage, or retroperitoneal hematoma, which requires ≥2 U of blood transfusion, causes significant hypotension, requires surgery, or decreases hemoglobin levels to >50 g/L. The secondary end points were MACE, MI, and cerebrovascular accident (CVA). The common variable definitions of primary and secondary outcomes are included in Table 1 ( Supplementary Section 3 ).
| Variable | Feldman 2013 | Kwok 2015 | Maynard 2013 | Pandie 2015 | Pristipino 2007 * | Rao 2008 | Gargiulo 2018 | Stehli 2020 | Jolly 2011 |
|---|---|---|---|---|---|---|---|---|---|
| Sample (n) TR/TF | 178,643/ 2,642,231 | 182,511/229,503 | 1163/22,436 | 3507/3514 | 299/601 | 7804/585,290 | 4197/4207 | 7949/8381 | 3507/3514 |
| Age, y | 63/64 | 65.5/66.2 | 65/65 | 63.1/62.7 | 66/67 | 64/65 | 67.3/67.2 | 63.7/66.6 | 62/62 |
| Male | 70.2/66.9 | 76.2/72.4 | 69/72 | 74.1/72.9 | – | 70.9/65.6 | 74.5/72.4 | 80/73.3 | 74.1/72.9 |
| BMI, kg/m 2 | 29.8/29.0 | – | – | – | 27/26 | 29.7/29.1 | 26.9/27 | 29.1/28.4 | – |
| CAD risk factors | |||||||||
| HTN | 81.5/80.2 | 52.5/50.8 | 78/75 | 60.4/59.1 | – | 79.5/78.6 | 62.5/63.8 | – | 60.4/59.1 |
| HLD | 79.8/77.9 | 56.6/56 | 80/75 | 42.9/44.9 | – | ||||
| DM | 35.6/35 | 17.8/18.2 | 38/31 | 22.3/20.5 | – | 33.1/34 | 22.8/22.4 | 18/21.8 | 22.3/20.5 |
| Smoker | 28.4/28.2 | 59.8/54.8 | 21/25 | 30.9/31.2 | – | – | 34.7/33.9 | – | 30.9/31.2 |
| Prior MI | 27.4/28.9 | 23.3/26.3 | 28/26 | 17.8/17.7 | – | – | 13.9/14.7 | – | 18.8/17.7 |
| Prior CABG | 8.9/19.3 | 4.7/10.9 | 11/17 | 2.3/2.1 | – | 10.4/20.8 | 2.6/3.5 | 1.2/10.1 | 2.3/2.1 |
| PAD | 11.6/11.8 | 4.7/4.4 | 15/9 | 2.6/2.3 | – | 15.8/12.6 | 8.1/8.8 | 2.4/4 | 2.6/2.3 |
| HD | 0.9/2.1 | – | – | – | – | – | 0.1/0.1 | – | – |
| Study characteristics | |||||||||
| Year | 2013 | 2015 | 2013 | 2015 | 2007 | 2008 | 2018 | 2020 | 2011 |
| Study design | retro obs | retro obs | retro obs | RCT | prospect obs | retro obs | RCT | prospect obs | RCT |
| Center | multi | multi | multi | multi | single | multi | multi | multi | multi |
⁎ Data is given only for female sex. HTN = hypertension; DM = diabetes mellitus; MI = myocardial infarction; CABG = coronary artery bypass graft; PAD = peripheral artery disease; HD = hemodialysis; BARC = bleeding academic research consortium; TLR = target lesion revascularization; RBC = red blood cells; Hgb = hemoglobin; MACE = major adverse cardiac events; MACCE = major adverse cardiac and cerebrovascular events; CABG = coronary artery bypass graft.
The unadjusted OR for the primary and secondary outcomes were calculated using the random-effect model, following the Cochran-Mantel Haenszel method. We reported the estimated effect size as a point estimate with a 95% confidence interval (CI). In-between study heterogeneity was assessed using the I 2 statistic test, ranging from no, low, moderate, and high degrees of heterogeneity at <25%, 25% to 50%, 50% to 75%, and >75%, respectively. Statistical significance was determined using the benchmark of a p <0.05 and a CI of 95%. We depicted publication bias graphically and numerically using the Egger’s Regression Test ( Supplementary Section 4 ). We performed meta-regression analysis to determine potential effect modifiers based on the random-effect model and linear regression. , Univariate meta-regression was performed on data on arms TR and TF for demographics (age, gender, BMI), comorbidities (hypertension, hyperlipidemia, diabetes mellitus, peripheral artery disease, hemodialysis, smoking), previous cardiac history (MI, coronary artery bypass graft) ( Supplementary Section 5 ). Statistical analysis was performed using Stata Statistical Software: Release 16. (StataCorp LLC, College Station, Texas).
Our initial comprehensive search identified a total of 7,571 articles. After excluding duplicate (281) and irrelevant (7,249) studies, 41 were reviewed in the full-text form. A total of 9 studies qualified for quantitative analysis. Three studies were RCTs, and 6 were observational (4 retrospectives, 2 prospective) studies. The Preferred Reporting Items for Systematic Reviews and Meta‐Analyses flow diagram is shown in Figure 1 .
Baseline demographics, relevant comorbidities, and study characteristics are presented in Table 1 . This meta-analysis report results from a total study population of 3,889,257 (389,580 [10%] in the TR arm and 3,499,677 patients [90%] in the TF arm) with a total of 2,633,729 or 68% men and 1,255,528 or 32% women. The male gender comprised 73% and 67% of the TR and TF arm, respectively. The average age was 64.4 and 65.1 years in the TR and TF groups, respectively. The most common co-morbidity was hypertension, present in 68% and 67% of patients. The follow-up period ranged from in-hospital to 30 days after procedure. The most common primary end points were the incidence of major bleeding and mortality. The Newcastle-Ottawa Scale assessment showed all studies with high quality (score 07/9). The Cochrane bias assessment showed a low risk of bias ( Supplementary Section 2 ).
The procedural characteristics, including procedure intervention, intraprocedural anticoagulation, and stent information, are presented in Table 2 . Most patients presented with stable angina (31% and 34%) or non–ST-segment elevation MI (48% and 46%) within the TR and TF groups. Unfractionated heparin was used in 55% and 42% of TR and TF cases, respectively. In addition, a drug-eluting stent was most commonly used, as seen in 53% of TR cases and 51% of TF cases.
| Variable | Feldman 2013 | Kwok 2015 | Maynard 2013 | Pandie 2015 | Pristipino 2007 * | Rao 2008 | Gargiulo 2018 | Stehli 2020 | Jolly 2011 |
|---|---|---|---|---|---|---|---|---|---|
| Clinical presentation, % | |||||||||
| Stable angina | 21.4/19.1 | 37.1/42.7 | – | – | 19/32 | 47.3/41.4 | – | – | – |
| Unstable angina | – | – | 45/32 † | 28.4/25.8 | 21/23 | – | – | 12.7/14.7 | 44.3/45.7 |
| NSTE ACS | 68/62 | 40.8/35.3 | 49/46 † | 44.3/45.7 | 18/26 | 49.7/54.5 | 52.3/52.2 | 47/43.4 | 28.5/25.8 |
| STE ACS | 10.6/18.9 | 22.2/19.6 | 6/21 | 27.2/28.5 | 14/14 | 3.2/4.1 | 47.6/47.7 | 40/41.6 | 27.2/28.5 |
| Procedure status, % | |||||||||
| Elective | 43.7/39.5 | – | – | – | 82/78 | – | – | – | – |
| Urgent | 44.6/39.4 | – | – | – | 22/19 ‡ | – | – | – | – |
| Emergent | 11.5/20.6 | – | – | – | 22/19 ‡ | – | – | – | – |
| Salvage | 0.1/0.4 | – | – | – | – | – | – | – | – |
| Intraprocedural medications, % | |||||||||
| Aspirin | 90.5/89.7 | – | – | 98.9/99.05 | – | – | 5.45/6 | – | 99.2/99.3 |
| Unfractionated heparin | 76.3/52 | – | 66/43 | 32.5/31.05 | – | 77.27/52.52 | 46.75/42.6 | – | 33.3/31.6 |
| LMWH | 11.6/12.9 | – | 4/5 | 51.85/51.45 | – | 15.7/16.41 | – | – | 51.5/51.8 |
| Bivalirudin | 45.5/52.1 | – | 50/58 | 2.1/3 | – | 13.76//39.62 | 39.35/40.15 | – | 2.2 /3.1 |
| Glycoprotein IIa/IIIa inhibitor | 26/33.5 | 23.45/22.35 | 16/26 | 23.3/22.35 | 42/37 | 36.95/39.83 | 12.2/11.2 | 12.75/22.75 | 25.3/24 ( * in STEMI: 34.5/31.1) |
| PCI characteristics, % | |||||||||
| Coronary angiography | – | – | 99.8/99.8 | – | – | – | – | 99.8/99.8 | |
| BMS | 20.4/23.6 | – | – | 65.1/69.55 | – | – | 26.2/24.7 | – | 65.3/69.1 |
| DES | 72.2/67.5 | – | – | 38.3/34.5 | – | – | 64.55/65.9 | – | 38.2/34.6 |
| Contrast volume, ml | 178/186 | – | – | – | 206/209 | – | – | – | – |
| Arterial sheath size, % | |||||||||
| ≤5 Fr | – | – | – | 15.8/7.6 | 5.8/5 § | – | – | – | 14.4/6.8 |
| 6 Fr | – | – | – | 76.6/79.8 | – | – | – | – | 77.4/80.2 |
| ≥7 Fr | – | – | – | 0.95/5.4 | – | – | – | – | 1/6 |
| Number of coronary arteries narrowed, % | |||||||||
| 0 | – | – | – | – | 48/46 | – | – | – | – |
| 1 | – | – | – | – | 38/35 ∥ | – | – | – | – |
| 2 | – | – | – | – | 38/35 ∥ | – | – | – | – |
| 3 | – | – | – | – | 15/19 | – | – | – | – |
| Procedural outcomes, % | |||||||||
| Access site crossover | – | – | – | – | – | 6.4/4.25 | – | 7.6/2 | |
| PCI completion | – | – | – | – | – | 76.35/76.85 | – | 95.4/95.2 | |
| Postprocedural TIMI Flow grade ≥3 | 60.3/51.8 | – | – | – | – | – | 95.6/95.65 | – | – |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
