SURGICAL INFECTIONS Volume 17, Number 3, 2016 ª Mary Ann Liebert, Inc. DOI: 10.1089/sur.2015.093
Impact of a Surgeon-Led Peripherally Inserted Central Venous Catheter Team on Peripherally Inserted Central Venous Catheter-Related Complications and Costs Luise I.M. Pernar,1 Lindsay L. Wolf,1 Anupamaa Seshadri,1 and Vihas Patel1,2
Abstract
Background: Peripherally inserted central venous catheters (PICCs) are popular for a broad range of indications. As with other forms of central access, PICC use can be associated with serious and potentially costly complications. In 2000, in response to the rising popularity of PICC use, a surgeon-led team was created to steward their placement. All requests were screened rigorously to ensure rational use. Our hypothesis was that creation of a dedicated PICC team would decrease inappropriate PICC placement, overall complication rates, and cost. Methods: The study was a retrospective review of prospectively collected data captured in the PICC teammaintained database between 2000 and 2013. The database was reviewed for PICC indications, reasons PICC requests were denied, and septic or thrombotic complications after PICC placement. To estimate cost savings, PICC supplies and each occurrence of blood stream infection (BSI) and thrombotic complication was assigned a cost on the basis of the available literature. Results: Between 2000 and 2013, 35,651 PICC placements were requested, of which 24,638 (69.1%) were approved, 22,157 (62.1%) immediately and 2,481 (6.9%) after initial refusal in view of further review of the indications. Most (95%) of the PICCs inserted were placed at the bedside within 1 d of approval. Blood stream infections occurred in 5.9% of patients and thrombosis in 2.7%. The attributable costs saved by declining placement of unnecessary PICCs, assuming the same proportions of patients would have developed a complication, could be as high as $5.4 million (M) in supplies, $7.77 M in avoided BSI and $2.25 M in avoided thrombotic complications, for a total savings of $15.44 M. Conclusions: The implementation of a surgeon-led PICC team had a significant impact on the placement rate, reducing cost by supplies foregone and complications avoided. Cost savings related to PICC placement alone should be considered as the definite cost savings because of the judicious allocation of resources.
T
potentially serious complications: Blood stream infection (BSI) and thrombosis [3]. Both of these events increase the cost of care significantly and have potentially life-threatening implications [4,5]. Evidence suggests a dedicated PICC team makes the placement of these devices more efficient and reduces the rate of complications [6]. In 2000, a surgeon-led dedicated PICC team was created at our institution as part of the Metabolic Support Service (MSS). This service aimed to place PICCs at the bedside rather than the interventional radiology (IR) suite, reducing equipment costs [7]. The PICC team also used a selective screening protocol, based on literature evidence and practice patterns, to determine if a PICC request was appropriate. The aim of this selective screening protocol was to
he use of peripherally inserted central venous catheters (PICCs) has increased dramatically since their introduction in the 1980s. At this time, PICCs have largely supplanted other central venous access catheters (CVCs) as the preferred route to administer intravenous solutions that are hypertonic, sclerosing, or irritating to small veins, such as total parenteral nutrition (TPN), chemotherapy, antibiotics, and inotropic medications. Ease of placement at the bedside by trained nurses or physician assistants has propelled PICC popularity [1]. Although PICCs have numerous advantages and are relatively easy to maintain, both in the hospital and in an outpatient setting, they are not necessarily associated with fewer complications than other forms of CVC [2]. There are two 1
Department of Surgery and 2Division of Metabolic Support, Brigham and Women’s Hospital, Boston, Massachusetts. Presented at the Thirty-fourth Annual Meeting of the Surgical Infection Society, Baltimore, Maryland, May 1–3, 2014.
1
2
PERNAR ET AL.
minimize placement of unnecessary PICCs, avoiding central access-related complications and promoting overall cost saving. Materials and Methods Study design
This study is a retrospective review of a prospectively maintained database of PICC consults requested and performed at Brigham and Women’s Hospital (BWH) from 2000 to 2013. This is a tertiary teaching hospital affiliated with Harvard Medical School (HMS), located in Boston, MA. The study was approved by the Institutional Review Board. PICC request screening protocol
In April 2000, a dedicated team was established within the MSS consisting of surgeons and midlevel providers to direct PICC placement [7]. In order to ensure efficiency, high quality, and safe patient care, a screening protocol prioritizing PICC placement for hospital discharge therapy was established that the midlevel providers, under the supervision of MSS surgeons, used to evaluate whether a PICC was requested for appropriate indications and if there were contraindications to PICC placement. Appropriate indications for PICC placement were the need for: Infusion of intravenous medications for more than three d and plans to discharge the patients with these medications; access for total parenteral nutrition (TPN); and access when placement of a centrally inserted CVC was deemed too risky, such as in patients with lung allografts. If access were required for long periods, primary teams were encouraged to pursue tunneled line placement. In addition to considering the appropriateness of a request for PICC placement, several additional factors were considered when screening PICC consults, including suspected or confirmed BSI, abnormal coagulation parameters, other diagnoses, patient location, and appropriateness of PICC compared with alternative peripheral or central access. If a patient had a fever within 24 h of a PICC request, the primary team was mandated to obtain blood cultures to evaluate for a BSI, and a PICC would not be placed until this blood culture was negative. If a positive blood culture had been documented previously, the primary team was required to obtain surveillance cultures until blood cultures were negative.
FIG. 1.
At BWH, a blood culture prediction rule is used by the microbiology laboratory when reporting results. This rule is based on a validated multivariable algorithm developed by Bates and Lee [8]. True-positive probabilities are determined using a clinical prediction model that classifies all preliminary and final results into four risk categories (low, medium, high, or very high). The algorithm uses the organism risk category, the time until a culture becomes positive, and other cultures with same organism drawn within five d. This system of classification has been used at BWH since before the inception of the PICC team, with the criteria remaining constant during the study period. The PICC team considered high and very high probability cultures when delaying or turning down PICC requests. Placement of a PICC was deemed unsafe if the platelet count was below 20,000/mm3 or the International Normalized Ratio (INR) was above 3. If either laboratory aberrance was encountered, the primary team was asked to correct the coagulopathy by administration of platelets, vitamin K, or fresh frozen plasma. If coagulation could not be corrected, the request for PICC placement was declined. As suggested by the National Kidney Foundation Dialysis Outcome Quality Initiative (NKF-DOQI), patients with chronic renal insufficiency or end-stage renal disease were not considered for PICC placement in order to preserve upper-extremity and central veins for potential future fistula access procedures [9,10]. Patients with malignant solid tumors tend to have greater complication rates with PICCs than with alternative central venous access, especially with long-term access [11,12]. Therefore, PICC requests for patients with malignant tumors were approved only if PICC duration would be short, whereas PICC requests would be declined and patients would be referred for central venous access placement if a long need for access was anticipated. Data suggest that BSI and thrombosis incidence are greater in the intensive care unit (ICU) setting when PICCs are used [13,14]. Therefore, requests to place PICCs in ICU patients were not approved, and insertion of a CVC was encouraged instead. Placement of a PICC also was declined if a patient could receive medications via a peripheral intravenous catheter (PIV) or if a CVC was already in place. The exception to this rule was if a patient required TPN, as this required a dedicated, clean line. The process is shown in Fig. 1.
Process for selection or denial of a peripherally inserted central venous catheter.
SURGICAL CATHETER TEAM AND COMPLICATIONS
3
Placement and monitoring of PICC
Placement and monitoring of PICC was performed per the protocol established with the inception of the PICC team [7]. If a request was approved, PICC placement was performed at the bedside using ultrasound guidance. If bedside placement was unsuccessful, the patient was referred to IR for placement of the PICC line. With the introduction of the dedicated service, a database was established for quality assurance. This database tracked the following items for each PICC request: Indication for PICC insertion, reason a PICC was not approved, the team that performed placement, how many attempts at placement were made, vein used, length of catheter, PICC tip conformation, and any difficulty during placement, such as tip malposition. Of note, although the PICC team placed the lines, the team did not oversee use after placement or the duration of use. Guidelines for the use of central lines were encouraged but not enforced by the PICC team. Complications, specifically BSI and thrombotic problems, were tracked by the BWH Center of Excellence using hospital discharge queries with International Classification of Diseases (ICD)-9-CM codes 999.31, 999.32, and 999.33 for catheter-related infection and 451.82, 451.83, and 451.84 for thrombotic complications. The Center also provided comparable figures for non-PICC CVC. Cost analysis
Each PICC placement was estimated to cost $491 for equipment and labor. We used available cost analyses to estimate the cost of each occurrence of BSI at $11,971 [15] and of thrombosis at $7,594 [16]. The salaries of the PICC team and the surgeons overseeing the PICC team were not taken into consideration in the analysis. The physician assistants, or their equivalents, who screened PICC requests and placed PICCs, would have been employed if the dedicated program did not exist as described. The surgeons overseeing the PICC team receive a stipend of approximately $12,000 annually. Data review and statistical analysis
Data regarding PICC indications, reasons for not approving PICC placement, location of procedure, as well as septic and thrombotic complications were analyzed. Results PICC placement
Between June 2000 and October 2013, 35,651 PICCs were requested. Of these, 22,157 (62.1%) were approved imme-
FIG. 2. Comparison of complications in patients with and without a peripherally inserted central venous catheter. diately, and an additional 2,481 (6.9%) were approved after further review of the indications. Overall, 24,638 of the requested PICCs (69.1%) were placed. When the PICC team was first established in 2000, 67.9% of accesses were inserted at the bedside by the PICC team, with 26.8% of placements performed by radiologists in the IR suite. By 2013, 97% of placements were performed by the PICC team and only 3% by radiologists. After the introduction of the PICC team, PICC placement typically occurred, and same-day placement was requested. The indication for PICC placement was predominantly antibiotic administration (61.1%), followed by access for hydration or provision of TPN (22.5%) and administration of chemotherapy (12.9%). The fewest PICCs were placed for maintenance purposes such as providing intravenous access in patients with difficult access or for blood sampling (3.5%). PICC disapproval
The main reason that a PICC request was declined was that placement was requested for administration of intravenous medications for fewer than five d (65%). Requests for a PICC sometimes were refused, as an alternative access was established already (9.5%) or patients had suspected or documented BSI, as evidenced by positive blood cultures (6.1%). Placement also was refused because a patient had current or anticipated dialysis needs (3.1%), the patient was in the ICU (2.8%) or did not have an indication for PICC placement; i.e. no TPN approval (2.6%), or had uncorrected coagulopathy (0.3%). Finally, several PICCs were not placed because the patient refused the procedure (0.8%).
Table 1. Cost Savings from Use of a Peripherally Inserted Central Venous Catheter Team
PICC equipment Blood stream infection Thrombosis Total (US$)
Individual cost (US$)
Incidence of complications (%)
Cost (US$) for all PICCs requested (n = 35,651)
Cost (US$) for all PICCs placed (n = 24638)
Cost (US$) savings for all PICCs not approved (n = 11,013)
491 11,971
5.9
17,504,641 25,179,909
12,097,258 17,401,548
5,407,383 7,778,361
7,594
2.7
7,309,810 49,994,360
5,051,726 34,550,533
2,258,084 15,443,827
4
The most frequent reason an initially refused PICC ultimately was approved was a need for PICC despite the short duration of the anticipated use (57.5% of all delayed approvals). Other reasons for delayed approval were resolution of BSI with negative blood cultures (13.1%) or resolution of coagulopathy (0.3%); decision to proceed despite alternative access (9.9%), ICU status (3.7%), or documented renal failure (1.2%); ultimate TPN approval (3.2%); patient deciding to allow PICC placement (0.9%); and miscellaneous reasons not otherwise captured. PICC complications
Blood stream infection was observed in 5.9% of patients who had a PICC placed. Thrombotic complications occurred in 2.7% of patients. There was minimal difference between PICC and other CVCs with regard to complications. The overall infection rate was 7.0% for non-PICC CVC, and the rate of thrombotic complications was 2.5%. Neither of these differences reached statistical significance (p = 0.26 and p = 0.43, respectively). These data are summarized in Fig. 2. Cost analysis
Cost savings attributable to the PICC screening protocol are based on the placements that were avoided. Assumed costs, as outlined above, were $491 for equipment, $11,971 for each episode of BSI [15], and $7,594 [16] for each thrombotic complication. By avoiding placement of 11,013 PICC lines, $5.4 million (M) was saved in equipment costs. Additionally, by eliminating the 5.9% risk of BSI and 2.7% risk of thrombosis that would have been associated with the placement of the declined PICC lines, additional cost savings of $7.77 M for avoided BSI and $2.25 M for avoided thrombotic complications were realized. In total, implementation of the PICC screening protocol resulted in theoretical cost savings of $15.44 M over the period examined. Table 1 summarizes these calculations. More than 50% of the cost savings can be attributed to avoiding BSI; 35% of the cost savings is attributable to unneeded equipment. Avoiding thrombotic complications contributed 14.6% in cost savings. Discussion
Centrally inserted central venous catheters used to be the mainstay access for patients requiring central access. However, their use carries a significant risk of serious complications and high economic costs. Since it was first used in the 1980s, the PICC gained popularity because of its presumed advantages over other central venous lines. The use of PICCs has grown rapidly, as PICC placement is a cost-effective alternative to CVC. However, significant complications are associated with PICC placement, including BSI and thrombosis [1–5,17]. Given the risks associated with PICC placement, there has been a call for research that examines its benefits and risks and that helps physicians understand current patterns and indications for PICC use. The team-based approach and screening protocol implemented by our surgeon-led PICC team has grown out of clinical practice. It is supported by the literature and represents an institutional effort toward rational PICC placement with the dual goals of greater patient safety and cost containment. This effort was set in motion as emerging evidence suggests that PICCs are not without significant complications
PERNAR ET AL.
and should not be used without critical appraisal of the needs and alternatives. We have shown here that, as a result of the implementation of the PICC team and the screening protocol, nearly one third of all requests for PICC insertion are turned down, the majority for the simple reason that the PICC is not needed. Complication rates associated with PICCs at our institution are 5.9% for BSI and 2.7% for thrombosis. The BSI rate is reported as bacteremia episodes per patient and for this study was aggregated over the entire time period, which may explain why the rate appears high compared with published estimates [2]. Certainly, the PICC team adheres to best practices and uses a central line insertion bundle, which includes hand hygiene, appropriate skin preparation, and employment of maximum sterile barriers. On the basis of conservative cost estimates for PICC placement and each instance of BSI and thrombosis, implementation of the PICC service and screening protocol has led to estimated cost savings of $15.44 M over 14 years, or $1.1 M annually. As the incidence of BSI and thrombotic complications is similar for PICC and non-PICC CVC, it is not true that PICC can be considered a safer alternative. The best way to avoid a complication associated with any form of central access is to avoid any kind of central access. The study has limitations, including that it is a single institution study, that complications were captured in the inpatient setting only, and that cost savings were not adjusted for economic realities over time. However, the large database makes the findings robust despite these limitations. Also, the screening protocol is not specific to our institution but could be exported readily to other settings. As an additional limitation, it is likely that some patients who were disqualified from receiving a PICC line would not have developed a complication; for instance, patients who were turned down because they did not require prolonged central access. In this case, cost savings attributable to avoidance of potential BSI or thrombosis complications may be overestimated. Cost savings related to PICC placement alone should be considered as the definite savings secondary to judicious allocation of resources through the implementation of the described PICC placement algorithm. Implementing a screening protocol to direct PICC placement can reduce the total number of such placements, potentially decreasing infectious and thrombotic complications and leading to significant healthcare cost savings. Author Disclosure Statement
The authors have no conflicts of interest to declare. References
1. Cardella JF, Cardella K, Bacci N, et al. Cumulative experience with 1,273 peripherally inserted central catheters at a single institution. J Vasc Interv Radiol 1996;7:5–13. 2. Safdar N, Maki DG. Risk of catheter-related blood stream infection with peripherally inserted central venous catheters used in hospitalized patients. Chest 2005;128:489–495. 3. Pikwer A, Akeson J, Lindgren S. Complications associated with peripheral or central routes for central venous cannulation. Anaesthesia 2012;67:65–71. 4. Chopra V, Flanders SA, Saint S. The problem with peripherally inserted central catheters. JAMA 2012;308:1527–1528.
SURGICAL CATHETER TEAM AND COMPLICATIONS
5. Chopra V, Anand S, Krein SL, et al. Bloodstream infection, venous thrombosis, and peripherally inserted central catheters: Reappraising the evidence. Am J Med 2012;125:733– 741. 6. Alexandrou E, Spencer TR, Frost SA, et al. Central venous catheter placement by advanced practice nurses demonstrates low procedural complication and infection rates: A report from 13 years of service. Crit Care Med 2014;42:536–543. 7. Robinson MK, Mogensen KM, Grudinskas GF, et al. Improved care and reduced costs for patients requiring peripherally inserted central catheters: The role of bedside ultrasound and a dedicated team. J Paren Enter Nutri JPEN 2005;29:374–379. 8. Bates DW, Lee TH. Rapid classification of positive blood cultures: Prospective validation of a multivariate algorithm. JAMA 1992;267:1962–1966. 9. National Kidney Foundation–Dialysis Outcomes Quality Initiative clinical practice guidelines for vascular access. Am J Kidney Dis 1997;30:S150–S191. 10. El Ters M, Schears GJ, Taler SJ, et al. Association between prior peripherally inserted central catheters and lack of functioning arteriovenous fistulas: A case-control study in hemodialysis patients. Am J Kidney Dis 2012;60:601–608. 11. Cheong K, Perry D, Karapetis C, Koczwara B. High rate of complications associated with peripherally inserted central venous catheters in patients with solid tumours. Intern Med J 2004;34:234–238. 12. Snelling R, Jones G, Figueredo A, Major P. Central venous catheters for infusion therapy in gastrointestinal cancer: A comparative study of tunnelled centrally placed catheters
5
13.
14.
15. 16.
17.
and peripherally inserted central catheters. J Intraven Nurs 2001;24:38–47. Ugas MA, Cho H, Trilling GM, et al. Central and peripheral venous lines-associated blood stream infections in the critically ill surgical patients. Ann Surg Innov Res 2012; 6:8. Wilson TJ, Stetler WR J, Fletcher JJ. Comparison of catheterrelated large vein thrombosis in centrally inserted versus peripherally inserted central venous lines in the neurological intensive care unit. Clin Neurol Neurosurg 2013;115:879–882. Ong A, Dysert K, Herbert C, et al. Trends in central lineassociated bloodstream infections in a trauma-surgical intensive care unit. Arch Surg 2011;146:302–307. Spyropoulos AC, Lin J. Direct medical costs of venous thromboembolism and subsequent hospital readmission rates: An administrative claims analysis from 30 managed care organizations. J Manag Care Pharm 2007;13:475–486. Johansson E, Hammarskjold F, Lundberg D, Arnlind MH. Advantages and disadvantages of peripherally inserted central venous catheters (PICC) compared to other central venous lines: A systematic review of the literature. Acta Oncol 2013;52:886–892.
Address correspondence to: Dr. Luise I.M. Pernar Brigham and Women’s Hospital 75 Francis Street Boston, MA 02115 E-mail: [email protected]
Impact of a Surgeon-Led Peripherally Inserted Central Venous Catheter Team on Peripherally Inserted Central Venous Catheter-Related Complications and Costs Luise I.M. Pernar,1 Lindsay L. Wolf,1 Anupamaa Seshadri,1 and Vihas Patel1,2
Abstract
Background: Peripherally inserted central venous catheters (PICCs) are popular for a broad range of indications. As with other forms of central access, PICC use can be associated with serious and potentially costly complications. In 2000, in response to the rising popularity of PICC use, a surgeon-led team was created to steward their placement. All requests were screened rigorously to ensure rational use. Our hypothesis was that creation of a dedicated PICC team would decrease inappropriate PICC placement, overall complication rates, and cost. Methods: The study was a retrospective review of prospectively collected data captured in the PICC teammaintained database between 2000 and 2013. The database was reviewed for PICC indications, reasons PICC requests were denied, and septic or thrombotic complications after PICC placement. To estimate cost savings, PICC supplies and each occurrence of blood stream infection (BSI) and thrombotic complication was assigned a cost on the basis of the available literature. Results: Between 2000 and 2013, 35,651 PICC placements were requested, of which 24,638 (69.1%) were approved, 22,157 (62.1%) immediately and 2,481 (6.9%) after initial refusal in view of further review of the indications. Most (95%) of the PICCs inserted were placed at the bedside within 1 d of approval. Blood stream infections occurred in 5.9% of patients and thrombosis in 2.7%. The attributable costs saved by declining placement of unnecessary PICCs, assuming the same proportions of patients would have developed a complication, could be as high as $5.4 million (M) in supplies, $7.77 M in avoided BSI and $2.25 M in avoided thrombotic complications, for a total savings of $15.44 M. Conclusions: The implementation of a surgeon-led PICC team had a significant impact on the placement rate, reducing cost by supplies foregone and complications avoided. Cost savings related to PICC placement alone should be considered as the definite cost savings because of the judicious allocation of resources.
T
potentially serious complications: Blood stream infection (BSI) and thrombosis [3]. Both of these events increase the cost of care significantly and have potentially life-threatening implications [4,5]. Evidence suggests a dedicated PICC team makes the placement of these devices more efficient and reduces the rate of complications [6]. In 2000, a surgeon-led dedicated PICC team was created at our institution as part of the Metabolic Support Service (MSS). This service aimed to place PICCs at the bedside rather than the interventional radiology (IR) suite, reducing equipment costs [7]. The PICC team also used a selective screening protocol, based on literature evidence and practice patterns, to determine if a PICC request was appropriate. The aim of this selective screening protocol was to
he use of peripherally inserted central venous catheters (PICCs) has increased dramatically since their introduction in the 1980s. At this time, PICCs have largely supplanted other central venous access catheters (CVCs) as the preferred route to administer intravenous solutions that are hypertonic, sclerosing, or irritating to small veins, such as total parenteral nutrition (TPN), chemotherapy, antibiotics, and inotropic medications. Ease of placement at the bedside by trained nurses or physician assistants has propelled PICC popularity [1]. Although PICCs have numerous advantages and are relatively easy to maintain, both in the hospital and in an outpatient setting, they are not necessarily associated with fewer complications than other forms of CVC [2]. There are two 1
Department of Surgery and 2Division of Metabolic Support, Brigham and Women’s Hospital, Boston, Massachusetts. Presented at the Thirty-fourth Annual Meeting of the Surgical Infection Society, Baltimore, Maryland, May 1–3, 2014.
1
2
PERNAR ET AL.
minimize placement of unnecessary PICCs, avoiding central access-related complications and promoting overall cost saving. Materials and Methods Study design
This study is a retrospective review of a prospectively maintained database of PICC consults requested and performed at Brigham and Women’s Hospital (BWH) from 2000 to 2013. This is a tertiary teaching hospital affiliated with Harvard Medical School (HMS), located in Boston, MA. The study was approved by the Institutional Review Board. PICC request screening protocol
In April 2000, a dedicated team was established within the MSS consisting of surgeons and midlevel providers to direct PICC placement [7]. In order to ensure efficiency, high quality, and safe patient care, a screening protocol prioritizing PICC placement for hospital discharge therapy was established that the midlevel providers, under the supervision of MSS surgeons, used to evaluate whether a PICC was requested for appropriate indications and if there were contraindications to PICC placement. Appropriate indications for PICC placement were the need for: Infusion of intravenous medications for more than three d and plans to discharge the patients with these medications; access for total parenteral nutrition (TPN); and access when placement of a centrally inserted CVC was deemed too risky, such as in patients with lung allografts. If access were required for long periods, primary teams were encouraged to pursue tunneled line placement. In addition to considering the appropriateness of a request for PICC placement, several additional factors were considered when screening PICC consults, including suspected or confirmed BSI, abnormal coagulation parameters, other diagnoses, patient location, and appropriateness of PICC compared with alternative peripheral or central access. If a patient had a fever within 24 h of a PICC request, the primary team was mandated to obtain blood cultures to evaluate for a BSI, and a PICC would not be placed until this blood culture was negative. If a positive blood culture had been documented previously, the primary team was required to obtain surveillance cultures until blood cultures were negative.
FIG. 1.
At BWH, a blood culture prediction rule is used by the microbiology laboratory when reporting results. This rule is based on a validated multivariable algorithm developed by Bates and Lee [8]. True-positive probabilities are determined using a clinical prediction model that classifies all preliminary and final results into four risk categories (low, medium, high, or very high). The algorithm uses the organism risk category, the time until a culture becomes positive, and other cultures with same organism drawn within five d. This system of classification has been used at BWH since before the inception of the PICC team, with the criteria remaining constant during the study period. The PICC team considered high and very high probability cultures when delaying or turning down PICC requests. Placement of a PICC was deemed unsafe if the platelet count was below 20,000/mm3 or the International Normalized Ratio (INR) was above 3. If either laboratory aberrance was encountered, the primary team was asked to correct the coagulopathy by administration of platelets, vitamin K, or fresh frozen plasma. If coagulation could not be corrected, the request for PICC placement was declined. As suggested by the National Kidney Foundation Dialysis Outcome Quality Initiative (NKF-DOQI), patients with chronic renal insufficiency or end-stage renal disease were not considered for PICC placement in order to preserve upper-extremity and central veins for potential future fistula access procedures [9,10]. Patients with malignant solid tumors tend to have greater complication rates with PICCs than with alternative central venous access, especially with long-term access [11,12]. Therefore, PICC requests for patients with malignant tumors were approved only if PICC duration would be short, whereas PICC requests would be declined and patients would be referred for central venous access placement if a long need for access was anticipated. Data suggest that BSI and thrombosis incidence are greater in the intensive care unit (ICU) setting when PICCs are used [13,14]. Therefore, requests to place PICCs in ICU patients were not approved, and insertion of a CVC was encouraged instead. Placement of a PICC also was declined if a patient could receive medications via a peripheral intravenous catheter (PIV) or if a CVC was already in place. The exception to this rule was if a patient required TPN, as this required a dedicated, clean line. The process is shown in Fig. 1.
Process for selection or denial of a peripherally inserted central venous catheter.
SURGICAL CATHETER TEAM AND COMPLICATIONS
3
Placement and monitoring of PICC
Placement and monitoring of PICC was performed per the protocol established with the inception of the PICC team [7]. If a request was approved, PICC placement was performed at the bedside using ultrasound guidance. If bedside placement was unsuccessful, the patient was referred to IR for placement of the PICC line. With the introduction of the dedicated service, a database was established for quality assurance. This database tracked the following items for each PICC request: Indication for PICC insertion, reason a PICC was not approved, the team that performed placement, how many attempts at placement were made, vein used, length of catheter, PICC tip conformation, and any difficulty during placement, such as tip malposition. Of note, although the PICC team placed the lines, the team did not oversee use after placement or the duration of use. Guidelines for the use of central lines were encouraged but not enforced by the PICC team. Complications, specifically BSI and thrombotic problems, were tracked by the BWH Center of Excellence using hospital discharge queries with International Classification of Diseases (ICD)-9-CM codes 999.31, 999.32, and 999.33 for catheter-related infection and 451.82, 451.83, and 451.84 for thrombotic complications. The Center also provided comparable figures for non-PICC CVC. Cost analysis
Each PICC placement was estimated to cost $491 for equipment and labor. We used available cost analyses to estimate the cost of each occurrence of BSI at $11,971 [15] and of thrombosis at $7,594 [16]. The salaries of the PICC team and the surgeons overseeing the PICC team were not taken into consideration in the analysis. The physician assistants, or their equivalents, who screened PICC requests and placed PICCs, would have been employed if the dedicated program did not exist as described. The surgeons overseeing the PICC team receive a stipend of approximately $12,000 annually. Data review and statistical analysis
Data regarding PICC indications, reasons for not approving PICC placement, location of procedure, as well as septic and thrombotic complications were analyzed. Results PICC placement
Between June 2000 and October 2013, 35,651 PICCs were requested. Of these, 22,157 (62.1%) were approved imme-
FIG. 2. Comparison of complications in patients with and without a peripherally inserted central venous catheter. diately, and an additional 2,481 (6.9%) were approved after further review of the indications. Overall, 24,638 of the requested PICCs (69.1%) were placed. When the PICC team was first established in 2000, 67.9% of accesses were inserted at the bedside by the PICC team, with 26.8% of placements performed by radiologists in the IR suite. By 2013, 97% of placements were performed by the PICC team and only 3% by radiologists. After the introduction of the PICC team, PICC placement typically occurred, and same-day placement was requested. The indication for PICC placement was predominantly antibiotic administration (61.1%), followed by access for hydration or provision of TPN (22.5%) and administration of chemotherapy (12.9%). The fewest PICCs were placed for maintenance purposes such as providing intravenous access in patients with difficult access or for blood sampling (3.5%). PICC disapproval
The main reason that a PICC request was declined was that placement was requested for administration of intravenous medications for fewer than five d (65%). Requests for a PICC sometimes were refused, as an alternative access was established already (9.5%) or patients had suspected or documented BSI, as evidenced by positive blood cultures (6.1%). Placement also was refused because a patient had current or anticipated dialysis needs (3.1%), the patient was in the ICU (2.8%) or did not have an indication for PICC placement; i.e. no TPN approval (2.6%), or had uncorrected coagulopathy (0.3%). Finally, several PICCs were not placed because the patient refused the procedure (0.8%).
Table 1. Cost Savings from Use of a Peripherally Inserted Central Venous Catheter Team
PICC equipment Blood stream infection Thrombosis Total (US$)
Individual cost (US$)
Incidence of complications (%)
Cost (US$) for all PICCs requested (n = 35,651)
Cost (US$) for all PICCs placed (n = 24638)
Cost (US$) savings for all PICCs not approved (n = 11,013)
491 11,971
5.9
17,504,641 25,179,909
12,097,258 17,401,548
5,407,383 7,778,361
7,594
2.7
7,309,810 49,994,360
5,051,726 34,550,533
2,258,084 15,443,827
4
The most frequent reason an initially refused PICC ultimately was approved was a need for PICC despite the short duration of the anticipated use (57.5% of all delayed approvals). Other reasons for delayed approval were resolution of BSI with negative blood cultures (13.1%) or resolution of coagulopathy (0.3%); decision to proceed despite alternative access (9.9%), ICU status (3.7%), or documented renal failure (1.2%); ultimate TPN approval (3.2%); patient deciding to allow PICC placement (0.9%); and miscellaneous reasons not otherwise captured. PICC complications
Blood stream infection was observed in 5.9% of patients who had a PICC placed. Thrombotic complications occurred in 2.7% of patients. There was minimal difference between PICC and other CVCs with regard to complications. The overall infection rate was 7.0% for non-PICC CVC, and the rate of thrombotic complications was 2.5%. Neither of these differences reached statistical significance (p = 0.26 and p = 0.43, respectively). These data are summarized in Fig. 2. Cost analysis
Cost savings attributable to the PICC screening protocol are based on the placements that were avoided. Assumed costs, as outlined above, were $491 for equipment, $11,971 for each episode of BSI [15], and $7,594 [16] for each thrombotic complication. By avoiding placement of 11,013 PICC lines, $5.4 million (M) was saved in equipment costs. Additionally, by eliminating the 5.9% risk of BSI and 2.7% risk of thrombosis that would have been associated with the placement of the declined PICC lines, additional cost savings of $7.77 M for avoided BSI and $2.25 M for avoided thrombotic complications were realized. In total, implementation of the PICC screening protocol resulted in theoretical cost savings of $15.44 M over the period examined. Table 1 summarizes these calculations. More than 50% of the cost savings can be attributed to avoiding BSI; 35% of the cost savings is attributable to unneeded equipment. Avoiding thrombotic complications contributed 14.6% in cost savings. Discussion
Centrally inserted central venous catheters used to be the mainstay access for patients requiring central access. However, their use carries a significant risk of serious complications and high economic costs. Since it was first used in the 1980s, the PICC gained popularity because of its presumed advantages over other central venous lines. The use of PICCs has grown rapidly, as PICC placement is a cost-effective alternative to CVC. However, significant complications are associated with PICC placement, including BSI and thrombosis [1–5,17]. Given the risks associated with PICC placement, there has been a call for research that examines its benefits and risks and that helps physicians understand current patterns and indications for PICC use. The team-based approach and screening protocol implemented by our surgeon-led PICC team has grown out of clinical practice. It is supported by the literature and represents an institutional effort toward rational PICC placement with the dual goals of greater patient safety and cost containment. This effort was set in motion as emerging evidence suggests that PICCs are not without significant complications
PERNAR ET AL.
and should not be used without critical appraisal of the needs and alternatives. We have shown here that, as a result of the implementation of the PICC team and the screening protocol, nearly one third of all requests for PICC insertion are turned down, the majority for the simple reason that the PICC is not needed. Complication rates associated with PICCs at our institution are 5.9% for BSI and 2.7% for thrombosis. The BSI rate is reported as bacteremia episodes per patient and for this study was aggregated over the entire time period, which may explain why the rate appears high compared with published estimates [2]. Certainly, the PICC team adheres to best practices and uses a central line insertion bundle, which includes hand hygiene, appropriate skin preparation, and employment of maximum sterile barriers. On the basis of conservative cost estimates for PICC placement and each instance of BSI and thrombosis, implementation of the PICC service and screening protocol has led to estimated cost savings of $15.44 M over 14 years, or $1.1 M annually. As the incidence of BSI and thrombotic complications is similar for PICC and non-PICC CVC, it is not true that PICC can be considered a safer alternative. The best way to avoid a complication associated with any form of central access is to avoid any kind of central access. The study has limitations, including that it is a single institution study, that complications were captured in the inpatient setting only, and that cost savings were not adjusted for economic realities over time. However, the large database makes the findings robust despite these limitations. Also, the screening protocol is not specific to our institution but could be exported readily to other settings. As an additional limitation, it is likely that some patients who were disqualified from receiving a PICC line would not have developed a complication; for instance, patients who were turned down because they did not require prolonged central access. In this case, cost savings attributable to avoidance of potential BSI or thrombosis complications may be overestimated. Cost savings related to PICC placement alone should be considered as the definite savings secondary to judicious allocation of resources through the implementation of the described PICC placement algorithm. Implementing a screening protocol to direct PICC placement can reduce the total number of such placements, potentially decreasing infectious and thrombotic complications and leading to significant healthcare cost savings. Author Disclosure Statement
The authors have no conflicts of interest to declare. References
1. Cardella JF, Cardella K, Bacci N, et al. Cumulative experience with 1,273 peripherally inserted central catheters at a single institution. J Vasc Interv Radiol 1996;7:5–13. 2. Safdar N, Maki DG. Risk of catheter-related blood stream infection with peripherally inserted central venous catheters used in hospitalized patients. Chest 2005;128:489–495. 3. Pikwer A, Akeson J, Lindgren S. Complications associated with peripheral or central routes for central venous cannulation. Anaesthesia 2012;67:65–71. 4. Chopra V, Flanders SA, Saint S. The problem with peripherally inserted central catheters. JAMA 2012;308:1527–1528.
SURGICAL CATHETER TEAM AND COMPLICATIONS
5. Chopra V, Anand S, Krein SL, et al. Bloodstream infection, venous thrombosis, and peripherally inserted central catheters: Reappraising the evidence. Am J Med 2012;125:733– 741. 6. Alexandrou E, Spencer TR, Frost SA, et al. Central venous catheter placement by advanced practice nurses demonstrates low procedural complication and infection rates: A report from 13 years of service. Crit Care Med 2014;42:536–543. 7. Robinson MK, Mogensen KM, Grudinskas GF, et al. Improved care and reduced costs for patients requiring peripherally inserted central catheters: The role of bedside ultrasound and a dedicated team. J Paren Enter Nutri JPEN 2005;29:374–379. 8. Bates DW, Lee TH. Rapid classification of positive blood cultures: Prospective validation of a multivariate algorithm. JAMA 1992;267:1962–1966. 9. National Kidney Foundation–Dialysis Outcomes Quality Initiative clinical practice guidelines for vascular access. Am J Kidney Dis 1997;30:S150–S191. 10. El Ters M, Schears GJ, Taler SJ, et al. Association between prior peripherally inserted central catheters and lack of functioning arteriovenous fistulas: A case-control study in hemodialysis patients. Am J Kidney Dis 2012;60:601–608. 11. Cheong K, Perry D, Karapetis C, Koczwara B. High rate of complications associated with peripherally inserted central venous catheters in patients with solid tumours. Intern Med J 2004;34:234–238. 12. Snelling R, Jones G, Figueredo A, Major P. Central venous catheters for infusion therapy in gastrointestinal cancer: A comparative study of tunnelled centrally placed catheters
5
13.
14.
15. 16.
17.
and peripherally inserted central catheters. J Intraven Nurs 2001;24:38–47. Ugas MA, Cho H, Trilling GM, et al. Central and peripheral venous lines-associated blood stream infections in the critically ill surgical patients. Ann Surg Innov Res 2012; 6:8. Wilson TJ, Stetler WR J, Fletcher JJ. Comparison of catheterrelated large vein thrombosis in centrally inserted versus peripherally inserted central venous lines in the neurological intensive care unit. Clin Neurol Neurosurg 2013;115:879–882. Ong A, Dysert K, Herbert C, et al. Trends in central lineassociated bloodstream infections in a trauma-surgical intensive care unit. Arch Surg 2011;146:302–307. Spyropoulos AC, Lin J. Direct medical costs of venous thromboembolism and subsequent hospital readmission rates: An administrative claims analysis from 30 managed care organizations. J Manag Care Pharm 2007;13:475–486. Johansson E, Hammarskjold F, Lundberg D, Arnlind MH. Advantages and disadvantages of peripherally inserted central venous catheters (PICC) compared to other central venous lines: A systematic review of the literature. Acta Oncol 2013;52:886–892.
Address correspondence to: Dr. Luise I.M. Pernar Brigham and Women’s Hospital 75 Francis Street Boston, MA 02115 E-mail: [email protected]
