Defining risk factors associated with difficult peripheral venous Cannulation: A systematic review and meta-analysis.

Abstract

Peripheral venous catheterization is a common technique in hospitals which is not always successful, resulting in multiple punctures and degradation of the vessels. This scenario, which we have termed 'difficult peripheral venous access', is associated to delays in care, obtention of samples or diagnosis, as well as a higher use of central catheters. This study intends to identify risk factors associated to the incidence of 'difficult peripheral venous access' in adults at hospital. We designed a systematic review of published studies (protocol PROSPERO 2018 CRD42018089160). We conducted structured electronic searches using key words and specific vocabulary, as well as directed searches in several databases. After validity analysis, we selected 7 studies with observational methodology. We found great variability in the definition of 'difficult peripheral venous access' and in the variables proposed as risk factors. Statistically significant factors through studies include demographic and anthropometric variables (gender, Body Mass Index), as well as medical and health conditions (diabetes, renal insufficiency, parenteral drug abuse, cancer chemotherapy), together with variables related to the vein or vascular access (vein visibility and palpability, vessel diameter, previous history of difficulty). Some studies have also considered variables related to the professional performing the technique. Meta-analyses were carried out for gender and obesity as potential risk factors. Only obesity appeared as a statistically significant risk factor with OR of 1.48; 95% CI (1.03 to 1.93; p = 0.016). Methodological heterogeneity prevented the development of further meta-analyses. It is essential to design future studies with diverse hospital populations, in which a wide selection of potential risk factors can be studied in a unique analysis. Our work identifies the most relevant variables that should be included in those studies.