Published trials comparing ORE from RRA and LRA were searched within the PubMed electronic database from database inception through June 5, 2025. The following keywords in the MESH major topics category were used for the search: “left-radial artery, right-radial artery, and operator radiation exposure.” Abstracts retrieved from the search terms were independently screened by two investigators for relevance and eligibility. All studies not written in English were excluded. Additionally, references from relevant articles were also reviewed for related studies. Relevant journal abstracts were reviewed after the time frame of the database search. Any disagreement between the two investigators for study inclusion was discussed and agreed upon by consensus. Inclusion criteria were: (1) randomized controlled trials or prospective observational studies (case-control or cohort studies); (2) enrolled patients allocated to LRA or RRA, and (3) recorded ORE at an anatomical location(s) (e.g., thorax, left eye, right eye, neck, wrist, and abdomen) with dosimeters located outside lead garments, and (4) used standard lead shielding arrangements in their cardiac catheterization laboratories. Exclusion criteria were: (1) anthropomorphic studies; (2) duplicate publication; (3) studies not peer reviewed; (4) studies that did not report pre-specified endpoints of interest; (5) studies that recorded aggregate ORE such as over a time period and not data on an individual case-by-case basis; or (6) ORE that was recorded with adjunctive radio-protection measures.

Baseline characteristics and relevant data were independently extracted from the individual studies through careful inspection of the complete article by two investigators. Any differences were resolved by consensus. Data extracted were year of publication, sample size, and procedural data of (1) ORE, demonstrated by cumulative radiation exposure (CR) in microsieverts (µSv), milliroentgen or millirem; and (2) normalized ORE expressed as cumulative radiation exposure/dose-area-product (CR/DAP). We extracted data from several anatomical locations. The thorax outcome consisted of studies reporting ORE at the thorax, sternum, left side of chest, trunk pocket, or left upper pocket. ^,,,,,,, The left eye outcome consisted of studies reporting ORE at the left eye, head to left of glasses, left side of the head, or head front in the middle between both eyes. ^,,,, The wrist outcome consisted of studies reporting ORE at the left wrist. ^{, 1415} The neck outcome consisted of studies reporting ORE at the neck or external thyroid. ^, The abdomen and right eye outcomes were from one study reporting ORE at those locations. Data with median and quartiles from these articles ^,,,, were converted into mean and standard deviation using the data estimation and conversion for meta-analysis (DECoMA) software. The article describing the DECoMA software provides the formulas used and references for where these formulas were obtained. The DECoMA article reports these formulas are validated approaches reported either in the Cochrane Handbook for Systematic Reviews of Interventions and/or Wan et al. ^, There were no special assumptions needed by the authors. All that was necessary was to input data into the DECoMA software. The specific DECoMA formula used to convert a median to a mean value was (0.7+ 0.39/sample size) (quartile1+quartile3)/2 + (0.3- 0.39/sample size) * median. The specific DECoMA formula used to convert quartile 1 and quartile 3 values into a standard deviation was (quartile3- quartile1)/[(2 * percentile of the standard normal distribution) (0.75*sample size– 0.125)/ (sample size + 0.25)].

Baseline patient characteristics extracted were age, sex, body mass index, hypertension, diabetes, dyslipidemia, and smoking status. Quality of studies were appraised according to the Jadad Scale for clinical trials or the Newcastle- Ottawa Quality Assessment Scale for cohort studies. This protocol was registered with the International Platform of Registered Systematic Review and Meta-Analysis Protocols (INPLASY) on August 29, 2025 (INPLASY registration number: INPLASY20258009).

The effect size of ratio of means with 95% confidence intervals were calculated. The formula used for the ratio of means was ln(LRAmean/RRAmean). The standard error formula used for calculating the confidence interval is the square root of (1/LRAsample size * (LRAstandard deviation/LRAmean) + 1/RRA sample size * (RRAstandard deviation/RRAmean) ). The ratio of means is exponentiated. The confidence interval is the ratio of means ± 1.96 * standard error which is then exponentiated . A random effect restricted maximum likelihood model was used. A number of heterogeneity measures were calculated consisting of Q, tau-squared, I ², and H ². Theta was analyzed to determine if the effect size significantly differed from 0. Small study effects indicating possible publication bias were analyzed with funnel plots, Begg’s test, and the Egger test. Forest plots were created to visualize the effect size. Stata SE Version 17 was used for all analyses and to create the Forest plots (Stata Corporation, College Station, TX, 2021).

No extramural funding was used to support this work. The authors are solely responsible for the design and conduct of this study, all study analyses, the drafting and editing of the paper and its final contents.