Inpatient venous access practices: PICC culture and the kidney patient

Introduction

Local and national fistula promotion initiatives have increased use of arteriovenous fistulas (AVFs), reducing vascular access infections and improving delivery of adequate dialysis (1-2-3). Central venous dialysis catheters promote a chronic inflammatory state, create an ongoing risk of bloodstream infection and have been shown to increase both mortality and costs of chronic hemodialysis (4-5-6).

In 2011, 387,914 individuals received hemodialysis in the United States (7). The number of patients on dialysis is expected to exceed 500,000 by 2020 (8), due to increasing prevalence of chronic kidney disease (CKD), diabetes, obesity and hypertension.

Mortality is an important competing outcome for patients with kidney disease. Although CKD may be diagnosed when the estimated glomerular filtration rate (eGFR) is <60 mL/min/1.73 m², few such patients ultimately start dialysis (9). Establishment of AVF in patients who never dialyze subjects patients to unnecessary surgical procedures with potential complications, whereas failure to construct a fistula in advance may result in months of dialysis via central venous catheters during which patients may suffer infections and vein injuries (10-11-12-13). Nephrologists must estimate the probability of future dialysis, refer appropriate candidates for fistula surgery and manage others more conservatively. While the oncoming need for dialysis can be quite evident in certain patients, and the high probability of mortality prior to kidney failure can be predicted easily in others, the final outcome may be uncertain for many kidney patients. Preservation of peripheral veins is advisable for patients with an uncertain renal prognosis.

Fistula promotion has been recognized as a national priority in both Healthy People 2010 and 2020. Both the American Society of Diagnostic and Interventional Nephrology (ASDIN) (13) and the National Kidney Foundation’s Kidney Disease Outcomes Quality Initiatives (KDOQI) (14) have published guidelines for the preservation of peripheral vessels in patients with CKD. Patients with stages 3-5 CKD, patients on dialysis and patients with kidney transplants should have peripheral veins protected. Venipuncture should ideally be restricted to dorsal hand veins, sparing forearm veins and the antecubital fossae. Central venous access is ideally placed in the internal or external jugular veins. Subclavian lines and peripherally inserted central catheters (PICCs) should be avoided. The Centers for Medicare and Medicaid Services has sponsored the Fistula First Breakthrough Initiative, which incorporates a position statement regarding PICC avoidance (available at http://esrdncc.org/ffcl/) and the Vascular Access Advisory Panel of ESRD Networks 9 and 10 has developed a PICC Avoidance Toolkit (available at http://www.therenalnetwork.org/home/form1.php).

These guidelines might be readily achievable, were CKD patients interacting with the healthcare system exclusively via their nephrologists. However, the number of patients with CKD greatly exceeds the number of nephrologists (15), and the numerous comorbid conditions that afflict patients with CKD ensure that they access the healthcare system via multiple routes. A hemodialysis patient is hospitalized a median of 1.84 times per year, for 11.7 days of inpatient care (16). Hospitalization of persons with CKD is a function of their comorbid diagnoses, but likely exceeds the number of dialysis patients admitted to hospitals on any given day.

The complex realities experienced by kidney patients admitted to American hospitals differ greatly from these guidelines. Hospitalization is potentially a time in which progress can be made toward a fistula, since the necessary diagnostic and surgical resources for fistula surgery are available without transportation issues. Patients and their families may be especially motivated to proceed with fistula surgery, during a hospitalization for dialysis catheter complications. However, very few vascular access creation procedures occur because reimbursement policies discourage the performance of imaging or surgery in the inpatient setting. Increasing requirements to justify the necessity of hospitalization, with orders for intravenous medications and daily laboratory monitoring, encourage the maintenance of continuous vascular access in inpatients. Patients who are unaware of the hazards of vein depletion (or unaware that they have kidney disease) may request to have their intravascular devices in their nondominant arms. Some infection control initiatives emphasize PICC as a measure to reduce central line bacteremia (17, 18). PICCs traverse the venous system all the way from the antecubital fossa to the great veins of the chest, and the geometry of their paths provides extensive potential to inflict contact injuries upon multiple veins, resulting in stenoses, thrombosis and atresia, all of which can prevent future fistula construction.

Our dialysis program has had a successful fistula promotion program that has sustained high rates of AVF in our prevalent population (3). We have noted increasing failures of initially successful upper arm fistulae due to stenoses and occlusions in proximal upper arm veins, in areas that are inaccessible to vascular devices other than PICC and midline catheters. Our quality assurance initiatives therefore studied the vascular access practices in the tertiary teaching hospital that serves our predialysis population.

Materials and Methods

In May 2010, the electronic medical records of all patients registered in our hospital were assessed, to identify all patients with eGFR <60 mL/min/1.73 m². These patients were classified into National Kidney Foundation CKD stages by eGFR_MDRD. Dialysis patients were classified into stage 5 CKD, regardless of eGFR. Each CKD patient was examined by an investigator, who assessed both arms for intravenous devices. Patients with eGFR ≥60 mL/min/1.73 m² were not assessed. Patients were asked which hand was used for writing; medical records and family members helped classify the dominant arm of patients who were sedated or nonverbal. The locations of any limb protection bracelets, central lines, dialysis fistulas, tunneled dialysis catheters or peritoneal dialysis catheters were noted. Examiners did not attempt to ascertain the locations of phlebotomy punctures or discontinued devices.

This survey was repeated 1 year later, with identical methodology. Between the two surveys of the hospital, we pursued an educational initiative aimed at increasing awareness of vein protection, and promoting the use of limb protection devices by hospital nephrologists and nephrology fellows. Interventions included lectures and workshops on Fistula First and vein protection, at both the departmental and training program level, including a Medical Grand Rounds presentation. Complications of failed vascular access were included in Morbidity & Mortality Conferences, and relevant articles were presented at our Journal Club. On the wards, assessment of vascular access was incorporated into the daily expectations for care plans. Fellows, rotating residents and attending staff were asked to carry limb protection bracelets at all times, and encouraged to apply them wherever needed.

In addition, the entire hospital population was surveyed as part of a separate quality assurance initiative in 2012, to determine the rates and distribution of PICC throughout the hospital. CKD stage and eGFR were not collected on patients without PICC. During this survey, the ordering practitioner was contacted, and asked to provide a rationale for ordering PICC placement.

Statistical analysis was performed using Minitab, version 16. Age was expressed as mean and standard deviation. Categorical variables were expressed as percentages, including gender, level of care and CKD stage. Group means were compared using a two-tailed Student’s t-test. Proportions were compared using chi-squared tests for homogeneity. Because eGFR was not collected in 2012, and PICC rate was not determined in patients with eGFR >59 mL/min/1.73 m² in 2011, CKD stages were compared between the PICC patients in 2012, vs. the summed overall hospital populations in 2010 and 2011. P values of 0.05 or less were used to determine statistical significance.

The individual study protocols were all approved by the Institutional Review Board (IRB) of Allegheny General Hospital, with study designs in conjunction with hospital’s Patient Safety Committee and the Vascular Access Committee of the Division of Nephrology. The IRB granted waivers of written informed consent to these quality assurance observations, although examiners obtained oral consent from all conscious patients.

Results

In 2010, 391 inpatients were assessed; in 2011, 415 patients were assessed. Baseline demographic characteristics of the hospital population were similar between both surveys, and are summarized in Table I. In 2010, bed assignments were divided into Intensive Care, Telemetry and Nonmonitored. In 2011 the hospital opened two additional “step-down” units that provided an additional level of care with noninvasive positive pressure ventilation, intravenous infusions and chronic ventilator support. The distribution of CKD to different levels of care is summarized in Table II. The distributions of intravenous devices and limb protection bracelets before and after the educational initiative are summarized in Table III, which shows a slight increase in limb protection, but no detectable changes in the distributions of PICC and other vascular access devices.

TABLE I -

Hospital demographics at the time of 2010 and 2011 surveys

Level of care	N	Male		Age (years)
		N	%	Mean	St Dev	P-value*
*P-value vs. nonmonitored.
2010
ICU	93	58	62.4	63.0	18.2	<0.01
Telemetry	110	53	48.2	67.6	12.1	<0.001
Nonmonitored	188	102	54.2	56.9	15.4
Total	391	213	54.4
2011
ICU	98	58	59.2	61.8	17.0	<0.01
Step-down	38	19	50.0	68.8	11.6	<0.001
Telemetry	98	47	48.0	66.5	15.2	<0.001
Nonmonitored	181	89	49.2	56	18.3
Total	415	213	51.0

TABLE II -

Distribution of ckd patients in the hospital, on the day of survey, 2010 and 2011

Level of Care	TotalN	CKDN	%CKD
2010
ICU	93	17	18.3
Telemetry	110	33	30.0
Nonmonitored	188	30	16.0
Total	391	80	20.4
2011
ICU	98	30	30.6
Step-down	38	17	44.7
Telemetry	98	28	28.5
Nonmonitored	181	33	18.2
Total	415	108	26.0

TABLE III -

Distributions of vascular access devices and limb protection bracelets, 2010 and 2011, before and after an educational initiative

	2010	2011
Clinical devices	N = 391	N = 415
Limb bracelet (n, %)	13 (3.3%)	19 (4.6%)
Device in protected arm (n)	0	2
PICC, nondominant arm (%)	12	13
PICC, dominant arm (%)	18	16
Peripheral IV, nondominant arm (%)	40	41
Peripheral IV, dominant arm (%)	28	38

The whole-hospital survey in 2012 included 375 inpatients and revealed an overall PICC rate of 22.9%; results are summarized in Table IV, which shows that patients with PICC were distributed preferentially in areas of higher inpatient acuity (P<0.001). PICC patients had lower eGFR than the general inpatient population, as defined by pooled data from the 2010 and 2011 surveys, which was statistically significant (P<0.01). The distribution of CKD stages is illustrated in Figure 1, which shows that PICC patients are less likely to have CKD stages 0-2, and more likely to have CKD stages 3-5. Ordering physicians reported that 42/86 PICCs were inserted for nonspecific indications such as “convenience,” or “poor vascular access.” Among specific indications, vancomycin (n = 15) and other antibiotics (n = 7) were the most common, along with vasopressors (n = 7), hypertonic saline (n = 6), milrinone (n = 4), chemotherapy (n = 3) and parenteral nutrition (n = 2).

TABLE IV -

Level of care, according to picc status in whole hospital survey, 2012

	PICC	Non-PICC
Level of care	%	%
ICU	19.7	12.0
Step-down	16.3	11.7
Telemetry	16.3	19.6
Nonmonitored	47.7	55.7
Total	100	100

Fig. 1 -

Discussion

Our observations show that vascular access practices applied to patients with abnormal kidney function are at considerable variance with guidelines. Increased nephrology awareness of vein protection did not affect the distribution of vascular access devices. PICCs were seen in more than 30% of patients with impaired kidney function, despite an overall PICC rate of 23%. Our data also show that patients with PICC have a higher probability of having eGFR <60 mL/min/1.73 m² than inpatients overall.

Expanded use of PICC has been enthusiastically accepted by practitioners and patients alike, and nephrologists may encounter considerable resistance to minimizing their use. Our analyses indicated that the rationales for insertion of a PICC were frequently nonspecific, and did not follow specific guidelines for the prudent use of central venous access.

Diabetes, cardiovascular disease and infections are all frequent in kidney patients, ensuring extensive exposure to hospitalization and vascular access devices. While PICCs represent short-term convenience to both practitioners and patients, indiscriminate use might easily be expected to have a significant negative impact upon the condition of upper arm veins and affect fistula prospects at the time of vascular surgery referral. Our observations suggest that there is very little recognition, outside of the nephrology community, of long-term negative consequences of PICC use in kidney patients.

Many CKD patients are not seen by nephrology teams in our hospital, which may explain why our educational initiative did not affect vascular access practices. Patients with eGFR 30-60 mL/min/1.73 m² were not regularly followed by nephrologists when they were hospitalized by other physicians. It is known that patients with CKD 3a and 3b get the majority of their outpatient care from primary physicians, but inpatient care of moderate CKD has not been well-described. The practice patterns in our hospital are typical of most community teaching hospitals of similar size. We cannot rule out the possibility that similar educational initiatives might be beneficial at institutions where nephrology physicians see an unusually high proportion of CKD patients, but absent a reliable method for increasing nephrologist exposure, our future educational efforts will target partners in the hospital community who order and place intravenous devices. We will need outreach and education of primary care physicians, house officers, physician-extenders and PICC nurses. Outside of our hospital, increasing recognition of the consequences of PICC has resulted in the American Society of Nephrology identifying PICC as an avoidable procedure in patients with impaired kidney function “… when there is even a modest likelihood that dialysis will be required in the future,” as part of the national “Choosing Wisely” Initiative (19).

This study has some significant limitations. The cross-sectional methodology of these surveys makes it impossible to distinguish chronic from acute impairment of kidney function. If the 2010 and 2011 surveys had assessed the PICC status of patients with eGFR ≥60 mL/min/1.73 m² and the 2012 PICC study had captured the eGFR of patients without PICC, we would have been able a make a far more relevant eGFR comparison of PICC patients to non-PICC patients, as opposed to a comparison of PICC patients vs. overall hospital patients. However, the effect of this limitation would be to bias our results toward the null, and therefore does not invalidate our conclusion that PICC patients have lower eGFR. Had it been feasible to record discontinued devices and laboratory blood draws, we could have more accurately described the venipuncture burden of our inpatients. Despite this limitation, we have documented that kidney patients sustain surprisingly heavy exposures to venous injuries.

The strength of this study is that it captures the normal working conditions of a large teaching hospital over several years, and presents evidence for a widely adopted but disturbing practice trend. Our data corroborate and expand upon prior work that has already shown a strong and independent association between a history of PICC insertion and lack of a functioning AVF (20). PICCs have already been shown to increase bacteremia and upper extremity deep venous thrombosis, the latter condition having been extremely rare prior to the explosive proliferation of PICC (21). Further work is needed to clarify why PICC utilization has expanded so rapidly and to develop effective strategies to reduce PICC utilization.

Conclusions

Increasing numbers of PICC are placed in patients with kidney disease, despite guidelines discouraging their use. Future difficulty achieving dialysis fistulas is a logical downstream consequence of this trend. Strategies to optimize vein preservation will need to enlist multiple hospital stakeholders.

Distribution of CKD stages differs between PICC patients and the general inpatient population. Patients with PICC are represented by gray bars; the entire hospital is represented by black bars (P<0.01).

• 1. Lynch JR.,Wasse H.,Armistead NC.,McClellan WM. Achieving the goal of the Fistula First breakthrough initiative for prevalent maintenance hemodialysis patients. Am J Kidney Dis 2011; 57: 78-89
• 2. Vassalotti JA.,Jennings WC.,Beathard GA. Fistula First Breakthrough Initiative Community Education Committee. Fistula first breakthrough initiative: targeting catheter last in fistula first. Semin Dial 2012; 25: 303-310
• 3. McGill RL.,Marcus RJ.,Healy DA. Long-term outcomes of a fistula initiative: sustaining “fistula culture”. J Vasc Access 2006; 7: 83-86
• 4. Wasse H.,Kutner N.,Zhang R.,Huang Y. Association of initial hemodialysis vascular access with patient-reported health status and quality of life. Clin J Am Soc Nephrol 2007; 2: 708-714
• 5. Banerjee T.,Kim SJ.,Astor B.,Shafi T.,Coresh J.,Powe NR. Vascular access type, inflammatory markers, and mortality in incident hemodialysis patients: The Choices for Healthy Outcomes in Caring for End-Stage Renal Disease (CHOICE) Study. Am J Kidney Dis 2014; 64: 954-961
• 6. Solid CA.,Carlin C. Timing of arteriovenous fistula placement and Medicare costs during dialysis initiation. Am J Nephrol 2012; 35: 498-508
• 7. USRDS 2013 Annual Data Report: Atlas of Chronic Kidney Disease and End-Stage Renal Disease in the United States. Bethesda, MD: National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, ; : p. 225-
• 8. Collins AJ.,Foley RN.,Gilbertson DT.,Chen SC. The state of chronic kidney disease, ESRD, and morbidity and mortality in the first year of dialysis. Clin J Am Soc Nephrol 2009; 4: S5-S11
• 9. Landray MJ.,Emberson JR.,Blackwell L. Prediction of ESRD and death among people with CKD: the Chronic Renal Impairment in Birmingham (CRIB) prospective cohort study. Am J Kidney Dis 2010; 56: 1082-1094
• 10. Demoulin N.,Beguin C.,Labriola L.,Jadoul M. Preparing renal replacement therapy in stage 4 CKD patients referred to nephrologists: a difficult balance between futility and insufficiency. A cohort study of 386 patients followed in Brussels. Nephrol Dial Transplant 2011; 26: 220-226
• 11. Oliver MJ.,Quinn RR.,Garg AX.,Kim SJ.,Wald R.,Paterson JM. Likelihood of starting dialysis after incident fistula creation. Clin J Am Soc Nephrol 2012; 7: 466-471
• 12. Conway B.,Webster A.,Ramsay G. Predicting mortality and uptake of renal replacement therapy in patients with stage 4 chronic kidney disease. Nephrol Dial Transplant 2009; 24: 1930-1937
• 13. Hoggard J.,Saad T.,Schon D.,Vesely TM.,Royer T. Vesely.,Royer T. Guidelines for venous access in patients with chronic kidney disease. Semin Dial 2008; 21: 186-191
• 14. KDOQI Clinical Practice Guidelines and Clinical Practice Recommendations for 2006 Updates: hemodialysis adequacy, peritoneal dialysis adequacy and vascular access. Am J Kidney Dis 2006; 48: S1-S322
• 15. Parker MG.,Ibrahim T.,Shaffer R.,Rosner MH.,Molitoris BA. The future nephrology workforce: will there be one? Clin J Am Soc Nephrol 2011; 6: 1501-1506
• 16. USRDS 2013 Annual Data Report: Atlas of Chronic Kidney Disease and End-Stage Renal Disease in the United States. Bethesda, MD: National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, ; : p. 240-
• 17. Al Raiy B.,Fakih MG.,Bryan-Nomides N. Peripherally inserted central venous catheters in the acute care setting: a safe alternative to high-risk short-term central venous catheters. Am J Infect Control 2010; 38: 149-153
• 18. Gunst M.,Matsushima K.,Vanek S.,Gunst R.,Shafi S.,Frankel H. Peripherally inserted central catheters may lower the incidence of catheter-related blood stream infections in patients in surgical intensive care units. Surg Infect (Larchmt) 2011; 12: 279-282
• 19. Williams AW.,Dwyer AC.,Eddy AA. American Society of Nephrology Quality, and Patient Safety Task Force. Critical and honest conversations: the evidence behind the “Choosing Wisely” campaign recommendations by the American Society of Nephrology. Clin J Am Soc Nephrol 2012; 7: 1664-1672
• 20. El Ters M.,Schears GJ.,Taler SJ. 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
• 21. Evans RS.,Sharp JH.,Linford LH. Risk of symptomatic DVT associated with peripherally inserted central catheters. Chest 2010; 138: 803-810