This study by Hasselberg et al. (2010) aimed to measure compliance with Swedish national guidelines addressing how long peripheral venous catheters (PVCs) should remain in situ, selection of the smallest possible PVC, and reporting of outcomes. The study included 196 patients (413 PVCs) from a single surgical ward. Data were collected from a local registry that was used to complement the original chart system. Professional staff were instructed to complete the registry correctly, for example, thrombophlebitis. Concerning relevance to clinical practice, Hasselberg et al. (2010) suggest that the findings could be easily implemented in clinical practice by revising the Swedish national guideline that PVCs should be removed every 12–24 hours to avoid complications, and instead recommend every 72 hours, as shown by the present study.
Apart from the useful findings on compliance with Swedish guidelines, it is also important to acknowledge the paper’s key limitations. The study was designed to describe compliance, but not to investigate different times between replacing in situ PVCs, an undertaking that would require a clinical controlled trial to yield solid evidence. The authors refer to a review by Idvall and Gunningberg (2006) that suggested using more clinical trials rather than descriptive studies to find evidence. A recently published Cochrane Library review (Webster et al. 2010) of clinically-indicated replacement vs. routine replacement of PVCs recommended that any future trial include at least 3000 subjects in each arm of a randomised controlled trial (RCT) to show true differences (e.g. reducing phlebitis from 9–7%). Applying a standard definition for phlebitis is also important. The Cochrane Library review was initiated because several observational studies over the years revealed uncertainty regarding ‘correct’ guidelines.
The conclusion of the present study – to consider revising the Swedish national guidelines – does not appear to be firmly grounded since it is based on findings from a descriptive registry study of around 200 patients from a single ward in one hospital. Several large descriptive studies present ‘evidence’ for different in situ times, but as the reviews of Idvall and Gunningberg (2006) and the Cochrane Library (Webster et al. 2010) assert, solid evidence would require RCTs. Moreover, the present study does not provide a definition of thrombophlebitis, rendering comparisons with other similar studies difficult. Although we agree with Hasselberg et al. (2010) that national guidelines should be based on evidence, the fact remains that their study does not contribute solid evidence for replacement times of in situ PVCs. It is essential to use the concept of evidence correctly, especially when we raise arguments concerning cost effectiveness, better use of nurses’ time, and saving patients from unnecessary suffering. Although descriptive studies are needed for quality improvement on different levels, we must be cautious about drawing conclusions and making clinical recommendations.