Improved Care and Reduced Costs for Patients Requiring Peripherally Inserted Central Catheters: the Role of Bedside Ultrasound and a Dedicated Team 
From the Departments of * Surgery and Correspondence: Malcolm K. Robinson, MD, Department of Surgery, Brigham and Women's Hospital, 75 Francis Street, Boston, MA 02115. Electronic mail may be sent to mkrobinson{at}partners.org. Presented in part at the third annual Nutrition Week, February 11, 2004. Background: We conducted a prospective quality assurance (QA) study to determine if a team dedicated to placing peripherally inserted central catheters (PICCs) would improve patient care and reduce costs. Methods: In April 2000, a dedicated team of physicians, physician assistants, nurses, and interventional radiologists (IR) was established to coordinate and approve all PICC placements at our hospital. Ultrasound (US) became available in November 2000 to assist with bedside PICC placement. A QA database was created allowing data from 3 time periods reflecting initiation of the PICC service (April–June 2000), initial implementation of bedside US-guided PICC placement (October–December 2000), and the current service (October–December 2002) to be analyzed and compared. Results: For all time periods analyzed, the PICC team found that one-third of PICC requests was inappropriate and, therefore, disapproved placement. With addition of US, the bedside PICC placement rate increased to 94% compared with 73% at service initiation. This was associated with an overall 80% decrease in average patient waiting time for a PICC, facilitating more timely discharges from the hospital. Finally, placement costs were reduced by 9% six months after initiation of our service and by 24% after US became available. Conclusions: A dedicated PICC team improves patient care by preventing inappropriate PICC placements and decreasing patient waiting times. A PICC team with US capability also reduces costs by minimizing expensive use of IR facilities and reducing hospital lengths of stay. A dedicated PICC service should become the standard of care for all hospitals with high-volume PICC use. Central venous access is often an essential part of patient care. This type of access is required for many different types of therapies, including chemotherapy, antibiotics, and parenteral nutrition (PN).1–3 Three main types of central catheters are available: percutaneously placed, nontunneled subclavian or jugular catheters; surgically placed tunneled and totally implantable catheters (eg, Hickmans®, Bard Access Systems, Salt Lake City, UT and PORT-A-CATHs®, Smith Medical MD, Inc., St. Paul, MN); and peripherally inserted central catheters (PICCs), which are rapidly rising in popularity.3 Although originally designed in the early 1900s, PICCs did not become widely used until the 1980s when improved catheter materials led to better patient tolerance. PICCs are now commonly used for intermediate- to long-term central venous access, with the ability to remain in place for up to 1 year.1,2 Many hospitals have experienced a rapid increase in demand for PICC placement because of the perceived benefit of PICCs over other central venous access devices (CVADs).4,5 Although there may be significant advantages to use of PICCs, there are also potential disadvantages. These include placing PICCs for inappropriate clinical indications, which may result in unnecessary patient complications. In addition, over-whelming demand for PICCs can lead to delays in patient care and discharge from the hospital if PICC placement is slow. Finally, there may be costly and inefficient use of hospital resources such as the interventional radiology (IR) suite. To address these concerns, our hospital implemented a dedicated PICC placement team to evaluate patients, approve PICC placement, and place PICCs when appropriate. An important addition to the PICC service armamentarium was portable ultrasound (US) to improve the success rate of bedside PICC placement. The aim of the present study was to assess the impact of the PICC team on delivery of patient care and the costs associated with PICC placement.
This prospective, quality-assurance study was conducted at the Brigham and Women's Hospital (BWH), a 716-bed tertiary teaching hospital of Harvard Medical School located in Boston, MA. Before April 2000, PICC placements at BWH were performed at the bedside by certified IV registered nurses (RNs) or by radiologists in the IR suite (Figure 1). Requests for PICC placement were made to either service at the discretion of the ordering physician. The RNs and radiologists screened patients primarily for clear contraindication to PICC placement. The RNs would refer patients to the IR suite if a PICC could not be successfully placed at the bedside.
In April 2000, a dedicated team was established to coordinate PICC placements at BWH. This team consists of surgeons, physician assistants (PAs), RNs, and radiologists. The team developed a standardized process for ordering PICC placements. In November 2000, the PAs were trained by interventional radiologists to perform line exchanges using a guidewire. In addition, the PAs were trained by a nurse practitioner to use a portable US machine designed specifically for assisting with vascular access (Site-Rite II Ultrasound System, Bard Access Systems, Salt Lake City, UT). The nurse practitioner also trained the PAs to place PICCs using the modified Seldinger technique. The PAs, with PICC physician supervision, were made responsible for screening PICC requests for both contraindications to PICC placement and clinical appropriateness. If deemed appropriate on both grounds, PICC placement was approved, and the PAs prioritized PICC placement according to clinical urgency, expected time of patient discharge, and order of receipt of PICC placement request. They then delegated PICC placement to other members of the team as appropriate. Usually, the RNs attempted PICC placement first. If the RN was unsuccessful or determined that placement would not be possible without US guidance, the PA attempted bedside PICC placement with US guidance (after November 2000). If the PAs were unsuccessful then the patient was referred to IR for PICC placement (Figure 2).
A database was developed with the initiation of the PICC team to monitor quality assurance (QA). The PAs are responsible for data entry for each patient in a prospective fashion. For this study, the following database queries were made: total PICCs placed, location of PICC placement (bedside vs IR), service placing PICC (IV nurse, PA, or IR), and length of time from PICC approval to PICC placement. Three 3-month time periods (ie, quarterly data) were chosen to reflect the evolution of the PICC service: PICC service inception (April–June 2000), initiation of bedside US use (October–December 2000), and the "current" service (October–December 2002). Cost for each PICC placement was calculated by determining the average cost for a nurse, PA, or radiologist to place a PICC according to nonphysician salaries, supplies, and cost of the IR suite if applicable. The 1-time cost of the US machine, approximately $8000, was not included in the cost of PICC placement. Cost of other materials required to place a PICC, such as US probe covers and echogenic needles, was included. Cost for placement of PICC in the IR suite did not include physician salaries or professional fees, because of the variability of insurance reimbursement for physician services. IR technician salaries were included in the "room cost" associated with use of the IR suite. Cost of routine chest radiograph after PICC placement was not included. In our institution, all patients require permanent radiographic confirmation of central venous access; thus, this cost is incurred for all patients receiving any type of central line, regardless of the service placing the catheter. For final calculations, we carried forward the cost of unsuccessful PICC placements. For example, when an RN failed to place a PICC and a PA was successful, the cost to place that PICC was the RN cost plus the PA cost. In the cases where a team member simply deferred an attempt to place a PICC and referred the patient to another member of the team, this cost was not included. The average costs for each service (RN, PA, or IR), including those failed attempts, were then used to determine the average cost per PICC placement for each of the 3 time periods under investigation.
Statistical analyses were done using analysis of variance (ANOVA) with post hoc Scheffé testing where appropriate (Statistica version 6.1 StatSoft, Tulsa, OK). When ANOVA was not appropriate, 95% confidence intervals were calculated to determine if statistically different from zero. A p value of < .05 was considered statistically significant.
During the time periods evaluated, 100% of the PICCs deemed appropriate for placement were successfully inserted by an RN, a PA, or a radiologist, if necessary. This was because the radiologists, using a combination of state-of-the art fluoroscopy and US in the IR suite, had a 100% success rate at placing PICCs when the IV nurses and PAs failed at the bedside. The number of PICC requests increased by 68% from PICC service inception to the current time, whereas the number of PICC placements increased by 75%. Despite the increase in PICC requests and placements, the percent of PICCs being disapproved remained unchanged at approximately 33% for all 3 time periods studied (Table I).
Success rate of bedside PICC placement significantly increased during the study. This resulted in a dramatic decrease in the need for radiologists to insert PICCs. There was a statistically significant decrease in IR use from service inception to the current service (27%–6%, 95% confidence interval [CI] for change, 0.02–0.22; p < .05) with a corresponding increase in the bedside placement success rate (73%–94%, Table II). This suggests that the use of US technology had a positive impact on the success rate of bedside PICC placement. Currently, IR is used only 6% of the time to place PICCs, representing a 78% drop in use of this costly service.
Time to PICC placement also improved with implementation of the dedicated PICC team (Table III). The combined bedside and IR placement time did not change from service inception to US initiation. For the entire PICC placement team, wait-time for PICC placement decreased by 80% with the implementation of US technology. This was largely because of the dramatic drop (88%) in IR placement time during this time interval, which was most likely a result of the decreased number of patients requiring PICC placement in the IR suite.
Cost for the IV nurses or PAs to place PICCs was similar but approximately 65% less than the cost for placement of a PICC in the IR suite (Table IV). Not surprisingly, the average cost per PICC placement for the 3 time periods dropped as the percent of PICCs inserted in IR decreased and the percent inserted at the bedside increased (Table V). The average cost per PICC decreased by 9% from service inception to US initiation and by 24% from service inception to the current service. Of note, the institution of US at the bedside was associated with a statistically significant decrease in average cost from service inception to the current service (p < .05).
The ease of PICC insertion and the low risk of severe complications such as pneumothorax and serious bleeding are significant advantages for use of this catheter over other CVADs.1,2 In addition, the cost associated with PICC insertion may be <10% of that of surgically placed CVADs.3 This has made PICCs the CVAD of choice in many clinical settings and has led to a rapid rise in their use at our hospital and others around the country. Although quite popular, there are several QA issues regarding PICCs that must be considered, especially when used in high volume. These issues are the focus of the present study. The first issue is whether PICCs are being used appropriately. Ease of PICC insertion at a relatively low cost can lead to "knee-jerk" use without careful consideration of the clinical appropriateness of the PICC. We found that request for PICC placement was inappropriate one-third of the time. One might expect that inappropriate PICC requests would decrease as experience with PICCs increased at our institution. However, the PICC denial rate remains unchanged despite a 75% increase in the number of PICCs being appropriately used. The most common reason for disapproving a PICC request was that central venous access was unnecessary and a peripheral IV was adequate and working well for the intended therapy. Other reasons for PICC denial included request for PN administration when PN was not indicated or appropriate; request for long-term IV antibiotic therapy when the infectious disease service deemed oral antibiotics equally or more appropriate than IV antibiotics; and, surprisingly, the ordering physician did not realize that a CVAD (usually a PORT-A-CATH®) was already in place. As a teaching hospital, the house officers are the ordering physicians for most PICCs at BWH. Unfortunately, they are often unfamiliar with the appropriate clinical indications for PICCs, a problem that renews itself every July with the arrival of new interns. We are attempting to close this knowledge gap through education sessions and note that this may be less of a problem in clinical settings where the ordering physicians are more familiar with indications for PICC insertion. However, this study strongly suggests that in institutions where there is high-volume PICC use, there should be screening for both contraindication to PICC placement and clinical appropriateness. Because PICCs are not without their complications, good patient care demands that PICCs only be used when truly indicated. A second QA issue related to PICC use is delays in patient care and discharge from the hospital when PICC demand exceeds timely insertion. Before formation of the dedicated PICC service, patients could wait up to 7 days for a PICC, and occasionally 10 days, at our hospital. Analysis of the situation indicated that delays were due to the high volume of requests over-whelming both the IV nurses and radiologists. Delays in PICC placement were especially long when placement in the IR suite was deemed necessary. The radiologists were often performing more urgent or emergency procedures for life-threatening conditions, and PICC placements were "at the end of the line." The high number of PICCs being placed unnecessarily exacerbated this problem for both the nurses and the radiologists. Establishment of the dedicated PICC team decreased the number of inappropriate PICC placements by screening for clinical appropriateness and shifted PICC placements from IR to the bedside. These changes were associated with drastic reductions in PICC placement time, delays in patient care, and delayed discharge from the hospital. Clearly, use of bedside US technology was a key to the success of the PICC service. Initiation of bedside US to guide PICC placements was associated with an 80% decrease in waiting time. Two meta-analyses6,7 assessing effectiveness of bedside US for placement of CVADs have shown improved placement success for first and overall attempts at catheter insertion compared with visual inspection, identification of anatomical landmarks, or palpation for suitable veins. We found that use of US improved our bedside success rate from 73% to 94%, which is consistent with the findings of others.8,9 Bedside PICC placement offers many other advantages such as avoiding a trip to the operating room or IR suite, which is associated with much patient apprehension. In addition, because other practitioners can place a PICC at the bedside, the patient does not need to wait for a physician to become available, or even worse, experience the frustration of being "bumped" for a more urgent procedure. Finally, limiting the volume of PICCs that needed to be done in IR decreased IR PICC placement time by 87%, making this a time-effective option when needed. Use of IR offers a near 100% success rate for PICC placement and is now readily available when needed if bedside placement attempts fail.5 Overall, 90% of our PICCs are placed on the same day of approval, which has increased from 71% when we started the service. The final QA issue addressed by this study is cost containment for PICC placement. Placement of a PICC in the IR suite is significantly more expensive than having it placed at the bedside by a nurse or PA. We found that the cost for IR placement is 65% higher than that for the PA and RN, even though the cost for PICC insertion in the IR suite did not include physician salaries. The cost we calculated for IR placement of a PICC is similar to costs reported by others.10 For example, Royer found that IR placement of a PICC cost $978 per PICC, which included physician time. Hence, the key to cost containment is to minimize the need for IR placement of PICC and shift as many PICC placements to the bedside as possible. The typical PICC placement team consists of certified PICC nurses who place PICCs at the bedside, with patients being referred to IR if bedside placement is unsuccessful. The typical success rate for a nurse is between 65% and 82%, which is consistent with the RN PICC placement success rate at our hospital.2,8,11,12 Adding bedside US significantly improved bedside placement success rate, which was the key factor in decreasing PICC placement costs by 24% from service inception to the current service. Of note, the overall cost savings of our PICC service is a combination of not only reduced placement costs but also reductions in costs related to delays in patient discharge from the hospital. We estimate that the overall savings related to institution of the PICC service at our hospital according to both of these factors exceeds $950,000 per year. In conclusion, this study systematically examined the effect of a dedicated PICC team on patient care and PICC placement costs in a prospective fashion. We found that a dedicated team has great impact in both areas, which has significant implications regarding use of PICCs, especially in the hospital setting. We now believe that a comprehensive, multidisciplinary team dedicated to evaluating patients, approving PICC placement, and inserting PICCs when appropriate should become the "gold" standard of care in institutions using PICCs in high volumes. Received for publication June 22, 2004. Accepted for publication April 29, 2005.
Journal of Parenteral and Enteral Nutrition, Vol. 29, No. 5,
374-379 (2005) This article has been cited by other articles:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Top
MATERIALS AND METHODS
RESULTS
