Improving Quality of Vascular Access Care for Hemodialysis Patients

Improving Quality of Vascular Access Care for Hemodialysis Patients

Joint Commission Journal on Quality and Safety Performance Improvement Improving Quality of Vascular Access Care for Hemodialysis Patients Ties va...

235KB Sizes 2 Downloads 315 Views

Joint Commission

Journal on Quality and Safety

Performance Improvement

Improving Quality of Vascular Access Care for Hemodialysis Patients

Ties van Andringa de Kempenaer, MD Pieter ten Have, MD Jacques Oskam, PhD, MD

urvival of patients with end-stage renal disease (ESRD) has increased considerably because of improved care and technical progress. Treatment options for patients with ESRD are either dialysis or kidney transplantation. Unfortunately, transplantation programs cannot deal with the number of patients with ESRD, and time on waiting lists is increasing, sometimes to longer than 6 years. Consequently, patients depend on hemodialysis for a longer period, and vascular surgeons are faced with the need to provide patients with longterm vascular access (hemodialysis requires access to large vessels). Quality of vascular access for hemodialysis is largely determined by the patency (durability) of access types, the number of re-interventional procedures to keep accesses functional, and the rate of infectious complications. The most commonly used types of vascular access are, from most to least desirable, a native vein arterio-venous fistula (AVF) on the arm, a prosthetic graft (PG) AVF on the arm, and a catheter in the central venous system (CVC). Extensive research shows that construction of a native vein AVF on the arm is currently considered best practice1–10 (Level 2 evidence; that is, well-controlled trials without randomization result in consistent reporting of the same findings11). However, construction of a native vein AVF is technically not always possible. A PG on the arm is a good alternative in these instances. A PG has worse patencies in the long term, higher infection rates, and a two- or threefold higher re-intervention rate compared with a native vein AVF.1–10,12 Nevertheless, a PG is the second-best option for vascular access on the arm. Sometimes vascular access construction on the arm is technically not possible at all, as when the patient has

S

April 2003

Article-at-a-Glance Background: Because quality of care for patients with end-stage renal disease (ESRD) has improved, they require long-term vascular access for hemodialysis. Construction of a native vein arteriovenous fistula (AVF) on the arm is considered best practice; a prosthetic graft (PG) AVF on the arm is a good alternative, although insertion of a central venous catheter (CVC), the third choice, is sometimes necessary. A quality improvement project was initiated at the dialysis unit of Rijnland Hospital (The Netherlands) to improve quality of vascular access care. Methods: Seventy-four patients were treated from January 2001 through June 2002. The list of preferred access operations was adapted from evidence-based guidelines. The percentages of CVCs and PGs were chosen as quality indicators. Results: Twelve of 19 patients (34%) appeared to be using CVCs unnecessarily. Actions were taken, and the CVC indicator decreased by 11%. The PG indicator decreased gradually from 24% to 8%. Discussion: Reductions in the use of CVCs and PGs suggest that the vascular access improvement project resulted in improvement of long-term vascular access for hemodialysis patients. A considerable decrease in the use of PGs and CVCs was achieved in 2001. However, a decrease of CVCs to <20% has still not been realized, perhaps because new hemodialysis patients referred to the dialysis unit have already had CVCs inserted. Summary and conclusion: Considerable improvement, as reflected in the number of hemodialysis patients with CVCs or PGs, can be achieved with a minimum of costs.

Volume 29 Number 4

Copyright 2003 Joint Commission on Accreditation of Healthcare Organizations

191

Joint Commission

Journal on Quality and Safety

severe arteriosclerotic disease or has had previous access surgery, or when veins or arteries are not of sufficient diameter. Because hemodialysis is a necessary life-saving measure, vascular access has to be achieved by other means—by insertion of a CVC. A CVC can be used for hemodialysis immediately, but it is associated with a high incidence of venous thrombosis and infection. Infection of a CVC may result in septicemia, which is life threatening, especially for hemodialysis patients.12 It is also associated with lower blood flow rates, so that dialysis will take longer or may even be inadequate, and central venous stenosis. Moreover, thrombosis may occlude the central veins, with the result that outflow obstruction may prevent the proper maturation of an AVF on the ipsilateral arm. Given all the possible life-threatening dangers of a CVC, it is the consensus that its use for vascular access should be limited and that the “arm vein first” policy should be followed.13–16 The dialysis unit of Rijnland Hospital (Leiderdorp, The Netherlands), a 480-bed teaching hospital in the western part of The Netherlands, was audited by the Dutch Dialysis Group (www.nefro.nl) in October 1999. Several issues with improvement potential for the hemodialysis unit were identified. Complaints of nursing staff concerning vascular access quality received special attention. The main criticism was the lack of a quality system for access surgery. An ongoing project to improve long-term vascular access for patients undergoing hemodialysis was started in January 2001 at the initiative of a vascular surgeon [J.O.] at the Rijnland dialysis unit. He consulted with a staff member [P.H.] at the National Institute for Health Care Improvement (CBO) in establishing the improvement project. To support the project, we—two vascular surgeons [J.O., T.A.K.] and the staff member [P.H.]—adopted the rapid-change improvement model of Langley et al,17 which entails the following steps: ■ What is our goal? Improvement of long-term vascular access for hemodialysis patients. ■ How do we know a change is an improvement? We quantified the quality of our access care by defining measurable quality indicators. ■ What changes lead to improvement? The changes that were to be implemented were intended to improve the value of the indicators. The changes consisted of implementing evidence-based guidelines by rewriting

192

April 2003

the protocol, establishing a multidisciplinary access team, and empowering nursing staff. The effects of these changes were monitored through measurement of the defined quality indicators. The Plan-Do-Study-Act (PDSA) cycle plays a key role in the Langley et al method; Figure 1 (p 193) shows the actions taken for each phase of the PDSA cycle. This article describes the actions taken and the results for the project, with a focus on the role of quality indicators.

Method Patient Characteristics All patients were treated in the hemodialysis unit of Rijnland Hospital; the unit has a maximum capacity of 55 patients. Most of the patients who undergo hemodialysis are outpatients who visit the unit two to three times a week. A total of 74 patients—50 men and 24 women (Table 1, p 194)—were treated in the period of investigation (Jan 2001–Jun 2002). During this period 11 of these patients died, 3 continued hemodialysis at another dialysis center, and 5 received kidney transplants. The population appears to be similar to other dialysis populations.

Rewriting the Protocol for Vascular Access Guidelines on how to best perform renal replacement therapies are well defined in the National Kidney Foundation–Kidney/Disease Outcome Quality Initiative (NKF–K/DOQI) guidelines concerning vascular access, specifically Guideline 3, “Selection of Permanent Vascular Access and Order of Preference for Placement of AV Fistulae.”9,10 On the basis of this guideline we adapted the “arm vein first” policy and represented the ideal sequence of access operations on the arms as follows: ■ First choice: A native vein AVF on the nondominant wrist. ■ Second choice: A native vein AVF on the dominant wrist. ■ Third choice: The nondominant elbow. ■ Fourth choice: The dominant elbow. ■ Fifth choice: The use of PG material on the arm in the same sequence as for native vein AVFs. If an existing PG requires surgical revision, the policy is to convert a PG into a native vein AVF, if possible. As

Volume 29 Number 4

Copyright 2003 Joint Commission on Accreditation of Healthcare Organizations

Joint Commission

Journal on Quality and Safety

The Plan-Do-Study-Act (PDSA) Cycle as Implemented in the Dialysis Unit of Rijnland Hospital

Empowering Nursing Staff Nursing staff were empowered to schedule the meetings, set the agenda, preside over the meetings, and perform the measurements needed to monitor quality of performance.

Defining Quality Indicators After reviewing the K/DOQI guidelines concerning vascular access9,10 and other literature, we concluded that the percentage of CVCs and the percentage of PGs on the total of patent arm accesses in our hemodialysis population were appropriate indicators. Ideally, these percentages should be as low as possible and would reflect the quality of vascular access for hemodialysis in our hospital. The CVC indicator was defined as the percentage of CVCs used in all hemodialysis patients. The PG indicator was defined as the percentage of PGs on the arm used in all patients with patent arm access. Nursing staff obtained data on Figure 1. The figure shows the actions taken for each of the PDSA cycles. CVCs, central venous catheters. the hemodialysis population at the end of the month to provide indicator measurements for the multidisciplinary meeting. recommended by the (K/DOQI) guidelines concerning 9 Each month, they plotted the CVC and PG percentages vascular access, a CVC will be inserted only if no access on a run chart, and they observed and analyzed trends. sites on the arms are present, or, if arm access were already constructed, they were not sufficiently mature to be used for hemodialysis. Results

CVC Indicator Establishing a Multidisciplinary Access Team A multidisciplinary team for vascular access care in the dialysis unit began meeting in May 2000. Three nephrologists, three dialysis nurses, two interventional radiologists, and two vascular surgeons decided to schedule a monthly meeting. The vascular surgeon [J.O.] who took the initiative in organizing the team felt that although quality of access is largely determined by the surgeon’s decision making, a multidisciplinary team can address particular access problems.

April 2003

Data collection started in January 2001. For the first 5 months, the mean percentage for the CVC indicator was 34% (range, 32%–36%; Figure 2, p 195)—a percentage that the vascular access team, at its May 2001 meeting, agreed was unacceptably high. Measurement revealed that this percentage represented 19 patients. Because the precise causes for the high rate of CVC vascular access were not known, an analysis of the indications for a CVC was performed in June 2001 (Study phase).

Volume 29 Number 4

Copyright 2003 Joint Commission on Accreditation of Healthcare Organizations

193

Joint Commission

Journal on Quality and Safety

Table 1. Clinical Characteristics of 74 Patients Treated in the Dialysis Unit of Rijnland Hospital, January 2001–July 2002.

CVC decreased from 38% in July 2001 to 24% in August 2001. A mean percentage of CVCs of 23% remained staAge (years) Mean, 64 (range, 19–81) ble until June 2002 (range, 27%–21%). Male to female ratio 50 male to 24 female In conclusion, monitoring the CVC Renal disease: indicator helped to identify a major Renal vascular disease n = 34 (46%) area for improvement, leading to a Diabetic renal disease n = 15 (20%) better mode of access for more than 11% of patients after particular Chronic renal disease n = 14 (19%) improvement interventions (mean Polycystic renal disease n = 4 (5%) percentage of 34% for January-June Acute glomerulonephritic disease n = 4 (5%) 2001 versus 23% for August 2001-June Other renal disease n = 3 (4%) 2002). The decrease of 11% was signifKidney transplantation n = 5 icant.* Table 2 shows the results of Deceased n = 11 causal analyses for December 2001 and June 2002. The waiting list for arm access surgery was now eliminated. Nevertheless, Chart review revealed five important causes for because early referral for surgery was not always possible, the high CVC score (Table 2, p 196). For 6 of the 19 a CVC was still present in 5 patients with maturing arm patients, construction of an AVF on the arm was possiaccesses, which represented a departure from the protoble, but the patients had not been referred to vascular col. Analysis revealed that 3 of these patients were surgeons. In fact, these patients constituted a “hidden referred to the Rijnland dialysis unit with CVCs already waiting list” for arm access surgery. In retrospect, we inserted. The remaining 2 patients had cardiac conditions know that these patients should have been referred earthat prevented construction of AVFs. lier and should have had a higher priority for surgery. Because the course of ESRD is often predictable, surgery could have been planned earlier. PG Indicator Another interesting indication for CVC appeared to be Monitoring of the PG indicator was performed simultafear of needles. Although the 3 patients with this indication neously with monitoring of the CVC indicator. As can be were satisfied with their CVCs, they were unaware of the seen in Figure 3 (p 197), in January 2001 the value of the associated risks. If they had been informed about the risks PG indicator was 24%. The percentage of PGs decreased of catheters and thoroughly convinced of their adverse to 8% in June 2002, mostly because of implementation of effects, they might have agreed to have AVFs on the arm. the arm vein first policy, which indicates that approxiSurprisingly, the causal analysis revealed a total of 12 mately 16% fewer patients with AVFs on the arm were patients for whom CVC was provided in the absence of exposed to the disadvantages of an PG. The decrease in proper indication. Consequently, these patients faced the mean percentage of 23% for January–June 2001 versus unnecessary risks of severe complications, possibly even 9% for August 2001–June 2002 was significant.* death. As soon as these results were available, the nine patients for whom surgery was suitable were offered Discussion early appointments at the outpatient surgery clinic for Reductions in the use of CVCs and PGs suggest that the access operation (Act phase). Furthermore, a new inforvascular access improvement project has resulted in mation protocol was developed for new ESRD patients. improvement of long-term vascular access for hemoNephrologists were asked to refer new patients to the dialysis patients. The most important changes were the vascular surgeon earlier to reduce the use of CVCs. * The finding that 8 consecutive points were below the median was conEight of the nine patients underwent CVC surgery in sidered statistically significant (Gitlow H, et al: Tools and Methods for the Improvement of Quality. Homewood, IL: Irwin, 1989.) June and July 2001, and the percentage of patients with

194

April 2003

Volume 29 Number 4

Copyright 2003 Joint Commission on Accreditation of Healthcare Organizations

Joint Commission

Journal on Quality and Safety

Run Chart with Percentage of Central Venous Catheters (CVCs) Used for Hemodialysis in the Total Hemodialysis Population, January 2001–June 2002.

Figure 2. This run chart shows the percentage of patients undergoing hemodialysis (HD) for whom CVCs were used. The medians for the percentage of CVCs (24%) and total number of HD patients (52) are shown.

implementation of guidelines for best-practice access care and the establishment of a multidisciplinary team, both of which have been employed in other studies of quality improvement in vascular access.18,19 In addition to the specific care improvements, each member of the vascular access team agreed that the working atmosphere was far more positive than it had been before. Furthermore, cooperation between specialists was constructive and without conflicts. All professionals wanted to adhere to the best practice. The nurses involved in the project confirmed that they had indeed felt responsible for access care, instead of only watching from the sideline. In the course of conducting the improvement work, the vascular access team developed a local benchmark for the percentage of CVCs to serve as a goal for continuous improvement of access quality. In view of the fact that we observed that about 12 patients had CVCs without proper indication, the internal benchmark was set at 10%–15%. A considerable decrease in the use of PGs and CVCs was achieved in 2001. However, the percentage of CVCs remained unchanged until June 2002, and a decline to <20% has still not been realized despite the team’s efforts. New hemodialysis patients who are referred to our dialysis unit may already have had CVCs inserted, so

April 2003

that the choice of access method is beyond our control. Further improvement would involve efforts to address the chain of care outside our hospital and in collaboration with other hospitals in the region. Vascular access use in Europe and the United States shows large variation even when adjusted for patient characteristics.20 These results may imply that the choice of vascular access type is largely determined by individual preferences. Several issues in access surgery may explain the observed differences and should be considered when attempting to improve the quality of care. Preferential construction of a native vein AVF increases patencies and decreases the incidence of reinterventions.3-8,12–16,20,21 The outcome of surgery may be improved by employing preoperative sonographic mapping, which may help the surgeon identify the best vessels for native vein AVF construction.22 Perhaps even more important is the timing of referral of patients with ESRD to the vascular surgeon to ensure peripheral vascular access when commencing hemodialysis.13,15,23 In addition, aggressive monitoring of functioning of arm accesses can also improve patency. The enormous improvement potential of a dedicated full-time vascular access coordinator has already been shown.15,23 The coordinator can help maintain a stream-

Volume 29 Number 4

Copyright 2003 Joint Commission on Accreditation of Healthcare Organizations

195

Joint Commission

Journal on Quality and Safety

ing the data was specified and defined clearly. In terms of validity, Reason for CVC May 2001 (nn = 56) Dec 2001 (nn = 52) Jun 2002 (nn = 48) the CVC and PG indicaNo arm access sites 4 (7%) 5 (10%) 5 (10%) tors appear to meet the Maturing arm 3 (5%) 2 (4%) 3 (6%) standard of consistency. A accesses large body of literature Cardiac failure 3 (5%) 3 (5%) 2 (4%) prevents AVF conconfirms that the use of struction CVCs or PGs is associated Patients fear of 3 (5% ) 2 (4%) 0 with decreased quality of needles vascular access.1–9,12–16,19–21,25–28 Waiting list for 6 (11 %) 0 0 However, the indicators’ arm access surgery validity is not precisely Total CVC / HD 19 (34%) 12 (23%) 10 (21%) known, given that they were population used in just one hemodialy* HD, hemodialysis; AVF, arteriovenous fistula. sis population. Further research, such as in risk lined vascular access program and conduct surveillance of adjustment, in other populations undergoing hemodialysis functioning for patients with varied types of vascular will be important in validating the quality indicators for 15,23 access, resulting in timely referral of patients to surclinical practice. The K/DOQI evidence-based guidelines9 geons and improved access patency. For example, at provide criteria for assessment of validity of the indicators. University Hospital Groningen, the vascular access coordinator, a nurse practitioner who functions as a liaison nurse Summary and Conclusion for all access issues, knows the patients well and mainConsiderable improvement, as reflected in the number of tains direct contact with the vascular surgeons to solve hemodialysis patients with CVCs or PGs, can be achieved problems and schedule operations on a short-term basis. with a minimum of costs. We described a local change In this improvement project, the CVC and PG indicators process in which the care for about 74 patients was helped the vascular access team identify the extent of the improved. The PDSA improvement method served as a potential for improvement in care. As the project proceedreliable management tool and can be used by professioned, their role changed from facilitating the change process als with no experience with change management. The indito measuring aspects of outcome and performance. The cators proved to be reliable for use as internal indicators CVC indicator directed attention to the long-term risk of to support improvement processes in the local situation. CVCs. As a result, all team members wanted to reduce the Therefore, care providers and quality professionals elsepercentage of CVCs, and a significant decrease was indeed where may wish to adapt our improvement work to imachieved. The PG indicator was developed to challenge the prove long-term vascular access in their own settings. J vascular surgeons to adhere to the “arm vein first” policy, and a significant decrease of PGs was also achieved. Ties van Andringa de Kempenaer, MD, is Surgeon and Broadening the use of the CVC and PG indicators from Fellow in Vascular Surgery, University Hospital, Groningen, internal quality improvement to comparison between The Netherlands. Pieter ten Have, MD, is Physician and dialysis units or benchmarking would require that they Staff Member, National Institute for Health Care Improvement (CBO), Utrecht, The Netherlands. Jacques meet criteria for reliability and validity.24 The indicators’ Oskam, PhD, MD, is Vascular Surgeon, Department of reliability seems acceptable because both detected sigVascular Surgery, Rijnland Hospital, Leiderdorp, The nificant changes and were clinically relevant (in terms of Netherlands. Please address correspondence to Ties van the risks to which patients were subjected). The retrieval Andringa de Kempenaer, MD, [email protected]. of data was straightforward, and the process of monitor-

Table 2. Causal Analysis for Central Venous Catheter (CVC)*

196

April 2003

Volume 29 Number 4

Copyright 2003 Joint Commission on Accreditation of Healthcare Organizations

Joint Commission

Journal on Quality and Safety

Run Chart with Percentage of Prosthetic Graft (PG) Arteriovenous Fistulae Used for Hemodialysis for All Patients with Arm Access, January 2001–June 2002

Figure 3. This run chart shows percentage of the patients undergoing hemodialysis (HD) for whom PGs were used. The medians for the percentage of PGs (11%) and the total number of patients with arm access (37.5) are shown.

References 1. Hakim R, Himmelfarb J: Hemodialysis access failure. A call to action. Kidney Int 54:1029–1040, 1998. 2. Bay WH, Van Cleef S, Owens M: The hemodialysis access: Preferences and concerns of patients, dialysis nurses and technicians, and physicians. Am J Nephrol 18:379–383, 1998. 3. Gibson KD, et al: Vascular access survival and incidence of revisions: A comparison of prosthetic grafts, simple autogenous fistulas, and venous transposition fistulas from the United States Renal Data System Dialysis Morbidity and Mortality Study. J Vasc Surg 34:694–700, 2001. 4. Murphy GJ, White SA, Nicholson ML: Vascular access for hemodialysis. Br J Surg Oct 87:1300–1315, 2000. 5. Fan PY, Schwab SJ: Vascular access: Concepts for the 1990s. J Am Soc Nephrol 3:1–11, 1992. 6. Windus DW: Permanent vascular access: A nephrologist’s view. Am J Kidney Dis 21:457–471, 1993. 7. Suhocki PV, et al: Silastic cuffed catheters for hemodialysis vascular access: Thrombolytic and mechanical correction of malfunction. Am J Kidney Dis 28:379–386, 1996. 8. Schwab SJ: Assessing the adequacy of vascular access and its relationship to patient outcome. Am J Kidney Dis 24:316–320, 1994. 9. III. NKF-K/DOQI Clinical Practice Guidelines for Vascular Access: update 2000. Am J Kidney Dis 37(1 suppl 1):S137–S181, 2001. 10. National Kidney Foundation Kidney/Disease Outcome Quality Initiative (NKF-K/DOQI): Clinical Practice Guidelines,

April 2003

www.kidney.org/professionals/doqi/guidelines/doqiupva_i.html#doqi upva3 (last accessed Jan 22, 2003). 11. The periodic health examination. Canadian Task Force on the Periodic Health Examination. Can Med Assoc J 21:1193–1254, 1979. 12. Jaar BG, et al: Septicemia in diabetic hemodialysis patients: Comparison of incidence, risk factors, and mortality with nondiabetic hemodialysis patients. Am J Kidney Dis 2:282–292, 2000. 13. Astor BC, et al: Timing of nephrologist referral and arterio-venous access use: The CHOICE study. Am J Kidney Dis 3:494–501, 2001. 14. Besarab A, et al: Unraveling the realities of vascular access: The network 11 experience. Adv Ren Replace Ther (4 suppl 1):S65–S70, 2000. 15. Kalman PG, et al: A practical approach to vascular access for hemodialysis and predictors of success. J Vasc Surg 30:727–733, 1999. 16. Feldman HI, Kobrin S, Wasserstein A: Hemodialysis vascular access morbidity. J Am Soc Nephrol 7:523–535, 1996. 17. Langley GJ, Nolan KM, Nolan TW: The foundation of improvement. Quality Progress 27(6):81–86, 1994. 18. Bosserman G: Multidisciplinary management of vascular access devices. Oncol Nurs Forum 6:879–886, 1990. 19. Allon M, et al: A multidisciplinary approach to hemodialysis access: A prospective evaluation. Kidney Int 53:473–479, 1998. 20. Pisoni L, et al: Vascular access use in Europe and the United States: Results from the DOPPS. Kidney Int 61:305–316, 2002.

continued

Volume 29 Number 4

Copyright 2003 Joint Commission on Accreditation of Healthcare Organizations

197

Joint Commission

Journal on Quality and Safety

References, continued 21. Feldman HI, et al: Hemodialysis vascular access morbidity in the United States. Kidney Int 43:1091–1096, 1993. 22. Allon M, et al: Effect of preoperative sonographic mapping on vascular access outcomes in hemodialysis patients. Kidney Int 60:2013–2020, 2001. 23. Welch KA, et al: Establishing the vascular access coordinator: Breaking ground for better outcomes. Nephrol News Issues, Nov 12, 1998, pp 43–46. 24. Rhew DC, Goetz MB, Shekelle PG: Evaluating quality indicators for patients with community-acquired pneumonia. Jt Comm J Qual Improv 27:575–590, 2001.

198

April 2003

25. Besarab A, et al: Utility of intra-access pressure monitoring in detecting and correcting venous outlet stenoses prior to thrombosis. Kidney Int 47:1364–1373, 1995. 26. Carlson DM, et al: Hospitalization in dialysis patients. Mayo Clin Proc 59:769–775, 1984. 27. Mayers JD, et al: Vascular access surgery for maintenance hemodialysis: Variables in hospital stay. ASAIO J 38:113–115, 1992. 28. Combe C, et al: Dialysis Outcomes and Practice Patterns Study: Data on the use of central venous catheters in chronic hemodialysis. Nephrologie 22:379–384, 2001.

Volume 29 Number 4

Copyright 2003 Joint Commission on Accreditation of Healthcare Organizations