Mermel and colleagues [1] describe their experience with a polyurethane-elastomeric hydrogel composite catheter (Landmark, Menlo Care, Inc., Menlo Park, California) in hospitalized patients. At our institution, the intravenous team (which is responsible for placing and maintaining peripheral intravenous access) found increased mechanical problems associated with the Landmark midline catheter compared with other types of peripheral venous access. We therefore evaluated noninfectious risk associated with the Landmark midline catheter. From 1 January 1994 to 31 December 1995, we prospectively evaluated 206 Landmark catheters placed in 206 patients (1826 patient-days). All catheters were used for administration of parenteral antibiotics. One hundred seventy-five patients had been hospitalized, and the remainder received antimicrobial therapy at home. The mean duration of catheterization was 10 days (range, 1 to 54 days). Forty-eight of 206 catheters (23%) were removed because of catheter-associated complications, including phlebitis of the arm (17 catheters), clot formation in the catheter (11 catheters), and infiltration at the insertion site (3 catheters). Seventeen catheters were removed for other reasons, including catheter kinking at the insertion site and continued discomfort, in either the lower or upper arm, without clinical evidence of local infiltration or phlebitis. The subgroup of 31 patients who received intravenous antibiotic therapy at home had a lower overall frequency of complications (16%) than did the hospitalized group. Five of these patients required catheter removal: two for clot formation and three for phlebitis.

Unlike Mermel and colleagues, we observed no allergic reactions to midline catheters. However, because of the incidence of noninfectious complications, particularly in hospitalized patients, we elected to discontinue the use of midline catheters in our institution.