Almost one quarter of patients who receive a peripherally inserted central catheter (PICC) while in hospital have advanced chronic kidney disease (CKD), a large sample of hospitalized patients in Michigan has found. But use of PICCs should be avoided in these patients in order to preserve future vascular access, and indeed, is contraindicated in recommendations, the authors stress.
“Vascular access is critical for patients with CKD, who may require renal replacement therapy,” David Paje, MD, MPH, University of Michigan Medical School, Ann Harbor, and colleagues write in their article published online June 4 in the Annals of Internal Medicine.
“[Yet] despite guidelines that recommend against the use of PICCs in patients with CKD, we found that such practice is common and discordant with guidelines,” they add.
An arteriovenous fistula (AVF) is the preferred choice for long-term hemodialysis and is more likely to succeed if veins have not previously been compromised by the insertion of a vascular catheter such as a PICC, Paje and colleagues explain.
And in an accompanying editorial, Marcia Silver, MD, Case Western Reserve University School of Medicine at MetroHealth Medical Center in Cleveland, Ohio, reminds readers that the mortality rate among patients receiving dialysis via a catheter is between 1.6 to 2.5 times higher than among dialysis patients with internal access and no catheter.
“The higher rate of catheter use in US hemodialysis patients likely explains the observed higher mortality compared with that in other countries,” Silver notes.
Statistics such as these have given rise to the “Fistula First, Catheter Last” campaign among those whose main goal is to drive down catheter use and increase use of native vein fistulae in the United States, she notes.
Although these campaigns have led to substantial improvement in the use of native vein fistulae among prevalent hemodialysis patients, national rates of catheter use by incident hemodialysis patients have hardly changed, Silver points out. The most recent report from the United States Renal Data System (J Vasc Surg. 2008;48:2S-25S) still shows that up to 80% of incident dialysis patients are receiving dialysis through a catheter.
“PICCs damage veins needed for vascular access for dialysis [and] they reduce success rates for creation of preferred vascular access types,” she notes. And “guidelines recommended avoiding PICCs in patients with stage 3b and worse chronic kidney disease [CKD], and rates of complications are high,” Silver stresses.
She calls the study by Paje and colleagues “extraordinary, with so many case records reviewed by trained data abstractors,” and the results “reaffirm prior observations that too little attention has been given to vein protection guidelines for patients with CKD,” she emphasizes, going on to recommend a number of “vein saving” strategies that could easily be employed.
Almost a Third of Patients With CKD in ICU Had PICCs Placed
In their study, Paje and coauthors used patient-level data from a multi-institutional quality collaborative effort to evaluate PICC placement in patients with CKD stage 3b or worse, defined as an estimated glomerular filtration rate (eGFR) of less than 45 mL/min/1.73m2.
“The primary outcome of interest was the percentage of patients with an eGFR less than 45 mL/min/1.73m2 (stage 3b or greater CKD) among all patients who received PICCs,” the investigators note.
More CKD patients in the intensive care unit (ICU) had PICCs placed, at 32.1%, compared with 18.9% on the wards. Older patients were much more likely to receive a PICC in both the ICU and on the ward than younger patients, the researchers note.
However, 3.4% of patients on hemodialysis also had a PICC placed during their hospital stay.
Two thirds of patients with CKD had their PICC removed prior to hospital discharge, and almost 30% of them had the PICC in place for 5 days or fewer — an interval when alternative venous access devices are considered more appropriate, the investigators point out.
And most often, CKD patients received a PICC that was 5-French or larger, and multilumen PICCs were placed more frequently than single-lumen PICCs both in patients in the ICU and on the wards.
In the ICU, 30.9% of patients receiving PICCs had an eGFR less than 45 mL/min/1.73m2; the corresponding percentage in wards was 19.3%.
And among patients with an eGFR less than 45 mL/min/1.73m2, multilumen PICCs were placed more frequently than single-lumen PICCs.
“Our findings raise the question of why PICC use is so widespread in patients with CKD,” say Paje and coauthors.
They suggest a number of possible explanations and say that “failure to develop…’safety systems’ — especially for physicians who order these devices and vascular access nurses who most often insert PICCs — might have led to inadvertent placement.”
Hospitals Should Leverage Nurses and Nephrologists to Improve Care
The most common complication associated with PICC use in patients with CKD was catheter occlusion; patients in the ICU had higher catheter occlusion rates, at 14.8%, compared with those on the ward, at 9.3%.
However, major complications, including venous thromboembolism and infection, were not uncommon.
In patients with an eGFR less than 45 mL/min/1.73m2, PICC placement varied widely across hospitals within the consortium (interquartile range, 23.7% to 37.8% in ICUs and 12.8% to 23.7% in wards).
And as the authors already observed, most PICCs placed in patients with CKD were multi-lumen devices, “many of which did not have a documented indication that suggested a true need for the device.”
“Thus,” they add, “not only are PICCs potentially being used inappropriately in patients with CKD, but provider choices about device characteristics may further increase risk for adverse events,” they stress.
The American Society of Nephrology recommends clinicians consult with a nephrologist prior to placing a PICC in patients with CKD stage 3 to 5 — a recommendation most hospitals in the study could have complied with, given that 90% of them had a nephrologist on staff who could have helped with PICC-centered decisions, the investigators emphasize.
Silver also points out in her editorial that nurses admitting patients to hospital “used to identify those with CKD who needed vein protection.” But in the 1980s, “we doubled the patient loads of most US hospital nurses, and engagement with this issue decreased thereafter.”
Paje and colleagues also emphasize this aspect of care, noting: “Vascular access nurses are well positioned to serve in this role and in this study were less likely to place PICCs in patients with CKD.”
“Empowering nurse-led vascular access teams to define clinical needs for access, consider a patient’s clinical profile, and make recommendations for appropriate device choice may help improve decision making,” they write.
And earlier referral of patients at risk for CKD progression to nephrology also results in less incident catheter use, says Silver. Yet many are still referred late or become “crash starts” to dialysis. “Risk stratification tools could help select those who need referral,” she indicates.
Strategies to Protect the Veins Include “Blue Bracelets”
Silver goes on to recommend a number of vein protection strategies. Firstly, the avoidance of PICCs in patients with CKD for whom short-tunneled internal jugular catheters could serve as an alternative.
Secondly, the use of smaller volume blood draws for everyone by using a pediatric tube for phlebotomy needs would help.
And “electronic medical records should include tools for identifying patients who need vein-saving techniques,” Silver suggests.
Also, “blue bracelets (blue for veins) marked ‘SAVE ARM VEINS’ [should be worn to] remind patients and staff to use vein-saving technique,” she emphasizes.
Lastly, the Paje and colleagues’ data remind the medical community that any vein protection strategy depends on decisions made by non-nephrologists caring for patients with CKD, she notes.
“Hence, education and engagement of the general medical community are critically important to further progress,” Silver concludes.
Paje and Silver have reported no relevant financial relationships.