Patterns of Peripherally Inserted Central Catheter Use Among Hospitalized Patients in a Community Hospital

Background:

Peripherally inserted central catheters (PICCs) have become an increasingly attractive option for short‐term vascular access in hospitalized patients. PICC placement is a minimally invasive procedure and provides a convenient, dependable intravenous access. These advantages, however, demand increased vigilance to minimize avoidable infectious and thrombotic complications. Though symptomatic in fewer than 5% of patients, PICC‐associated thrombosis — which increases with longer catheter dwell times — is present in greater than 30% when evaluated by follow‐up venography or Doppler ultrasonography. Hospitalized patients with chronic kidney disease (CKD) represent a particularly vulnerable population because upper extremity or central vein thrombosis may limit future hemodialysis access options. We conducted an observational study to identify patterns of PICC use in a typical community hospital, including patients with CKD.

Methods:

A retrospective chart review was conducted at a 130‐bed community hospital of all inpatient PICC procedures during 2007. Cases for inclusion were identified by diagnosis code (ICD‐9 38.93) and procedure notes documenting peripheral insertion. Individual medical records were reviewed to determine the timing of catheter placement and removal and catheter‐related complications . The Abbreviated Modification of Diet in Renal Disease (MDRD) Study equation was used to estimate glomeruiar filtration rate (GFR) based on patient age, sex, and serum creatinine.

Results:

For 2007, 307 PICC procedures were identified for review. One hundred and sixty‐one PICCs were removed during the hospitalization (hospital‐only PICCs). Among these hospital‐only PICCs, 65 (40%) exceeded dwell times of 7 days, and 23 (14%) were 2 days or less. One hundred and fifteen hospital‐only PICCs (71%) were removed on the final day of hospitalization. Twenty‐five hospital‐only PICCs (16%) were removed prior to discharge because of clinically suspected PICC‐related infection or thrombosis. The estimated GFR was less than 30 mL/minute in 57 PICC recipients (19%).

Conclusions:

A significant proportion of PICCs placed in the hospital were for inpatient vascular access only, with the majority removed on the day of discharge. We observed a significant risk for infection or thrombosis. We also observed a relatively high proportion of patients with advanced renal insufficiency receiving PICCs, potentially jeopardizing future hemodialysis access options. More research is needed to see if these patterns reflect broader trends. Hospitalists and other inpatient providers should stay vigilant against unnecessary PICC use and dwell times while staying attuned to the unique risks to CKD patients.

Author Disclosure:

B. Erb, none; J. Stein, sanofi, research funding; B. Gartland, none; D. Tong, none.