Inter-arm blood pressure differences (IAD) are normally distributed, and normal ranges for systolic IADs around +/- 10mmHg have frequently been proposed. Some studies report higher mean blood pressure (BP) readings for right compared to left arms, but this is not consistent. We undertook a systematic review and meta-analysis to estimate the usual difference in BP between right and left arms, and to explore any effect of hand dominance on IAD.Design and method:
EmBase, Medline and CINAHL were searched to October 2015 using text terms for IAD. Journal collections, conference abstracts, grey literature, and reference lists were also checked. We included studies measuring BP simultaneously in both arms, seeking additional data from study authors where needed. Study quality was assessed using the Newcastle-Ottawa scale.Design and method:
Mean IAD (right minus left) was calculated for each study and data were pooled by inverse variance in random effects models. Heterogeneity was quantified using the I2 statistic and explored within subgroup analysis by study quality.Results:
We identified 1672 unique citations; 27 studies contributed to meta-analyses. Systolic BP was 1.0mmHg (95%CI 0.7 to 1.3) higher for right compared to left arms (27 studies, 14398 subjects). Statistical heterogeneity was lower for high quality (I2 = 14%) compared to low quality studies (I2 = 77%). Diastolic BP did not differ between arms (0.1mmHg, -0.2 to 0.4; 24 studies: 13266 subjects).Results:
For right handed subjects systolic BP was 1.3mmHg (1.0 to 1.7) higher in the right arm (I2 = 0%; 6 cohorts, 3301 subjects) but did not differ in left handed subjects (0.1mmHg, -0.8 to 1.0, I2 = 0%; 272 subjects; p = 0.02). A non-significant opposite trend was seen for diastolic BP: -0.3mmHg (1.3 to 0.7) right handed; 0.7mmHg (-0.6 to 2.1) left handed (p = 0.23).Conclusions:
Overall systolic BP is 1mmHg higher in the right compared to the left arm. Right handedness dominates the population masking a different pattern in left handed people. Arterial anatomy and asymmetric arm development probably explain the differences observed. The findings are relevant for BP measurement protocols favouring non-dominant arms, and for future studies of inter-arm difference.