Does this patient have chronic kidney disease requiring preparation for renal replacement therapy?

As chronic kidney disease (CKD) progresses, an essential component of care becomes educating and preparing patients for end stage renal disease (ESRD) and possible treatment options. This chapter focuses on issues surrounding treatment for advanced renal failure and steps toward preparation for renal replacement therapy (RRT). The initial question for many providers is whether their patient is at risk for needing RRT. Over the last 10 years, multiple clinical guidelines have been published to address this issue.

Clinical guidelines

The National Kidney Foundation / Kidney Disease Outcome Quality Initiative (NKF/KDOQI) recommends educating patients about kidney failure and treatment options once renal function has declined below an estimated glomerular filtration rate (eGFR) of 30 ml/min/1.73 m, classified as stage 4. This approach is supported by other expert bodies including CARI (Caring for Australians with Renal Impairment), CSN (Canadian Society of Nephrology), ERA-EDTA (European Renal Association – European Dialysis and Transplant Association), United Kingdom Renal Association, and the RPA (Renal Physicians Association).

The broad educational goals include informing patients about complications, prognosis, and treatment options for advancing disease. Conservative therapy, transplantation, in-center hemodialysis, home and nocturnal hemodialysis, and peritoneal dialysis modalities should be addressed with appropriate patients as possible treatment choices.

Continue Reading

Along with educating patients about treatment options, patient care should include a plan for delaying disease progression, limiting symptoms of ESRD, timely dialysis access placement, and transplantation referral. Late referral to nephrology care, however, poses a significant barrier to appropriate RRT preparation.

Timing of nephrology referral

There is no consensus about the appropriate timing of referral for CKD. At minimum, patients with stage 4 or rapidly deteriorating stage 3 CKD should have prompt nephrology evaluation.

Though the incidence of late referral varies by the population studied, it is a widespread issue in renal care globally. In the United States Renal Data System (USRDS) Dialysis Morbidity and Mortality Study – Wave II, only 39% of patients undergoing hemodialysis had seen a nephrologist greater than one year before starting RRT. Furthermore, 25% of patients undergoing hemodialysis and 16% of patients undergoing peritoneal dialysis were initially evaluated less than 1 month before starting RRT. Overall 33% of patients were referred less than 4 months prior to initiating RRT.

This problem is not limited to the United States. Up to 30% of European patients with diabetes are referred within 1 month of initiating RRT, and data from the Australia and New Zealand Transplant Registry (ANZDATA) show that 26% of patients are referred less than 2 months before starting dialysis. Patients referred late are more likely to be from underrepresented groups, uninsured, or have a high burden of coexistent disease.

Although there are no prospective, randomized studies, the abundance of literature shows that late referral is associated with greater morbidity and mortality, and with unplanned starts to dialysis. Furthermore, unplanned start to dialysis is associated with higher mortality, high rate of catheter use for access, and higher rates of access related admissions.

Not surprisingly, modality selection is also effected by timing of referral. Kinchen et al found that patients referred to nephrology greater than 1 year before RRT initiation were more likely to choose peritoneal dialysis than those referred less than 1 year. It has also been shown that patients referred later are less likely to be waitlisted or offered renal transplantation prior to ESRD status.

Besides being ill-prepared for RRT, patients who present late are more likely to have suboptimal care of renal comorbid conditions. Patients referred late are more likely to have hypoalbuminemia and severe anemia, and they were less likely to have received erythropoietin or have a permanent access.

Predicting progression to ESRD

Since the publication of the NKF/KDOQI guidelines on classification of CKD and routine reporting of eGFR by laboratories, CKD awareness has improved.

The prevalence of CKD in the United States is estimated at 15.2% (NHANES 2003-2006) with 7.8 % of the population classified as having stage 3 and 0.5% of the population classified as having stage 4 to 5 CKD. Approximately 110,000 to 120,000 patients initiate RRT each year in the United States, and the overall ESRD prevalence is 550,000 to 600,000 patients.

Most patients with stage 3 and 4 CKD will never progress to ESRD. Preparing this entire group would strain limited resources and subjects many patients to unnecessary emotional duress and procedures.

As previously noted, inadequate preparation exposes those who are destined for RRT to avoidable morbidity and mortality. The preparatory process is time intensive and requires significant healthcare utilization. Hence, accurate prediction of those who are likely to need RRT is paramount. Recent literature has focused on models to aid in patient selection. Can we predict who will need RRT?

There is limited publication on the stage specific risk of reaching ESRD. It appears, that only a minority with stage 2-4 CKD will ever require RRT. In one of the more recent studies, Keith and colleagues found that 1.1%, 1.3%, and 19.9% of patients in stage 2, 3, and 4 CKD required RRT after 5 years of follow-up whereas the rate of death in these groups was 19.5%, 24.3%, and 45.7% over the same period (Kaiser Permanente Northwest Division database).

Progression to ESRD depends on population characteristics. Age in particular is a significant modifier of risk. Multiple studies show an increased risk of death compared to ESRD in the elderly and proportionally slower rate of GFR decline when compared to younger populations. For patients greater than 75 years of age, the risk of death exceeded ESRD for all levels of eGFR. The elderly have increased competing risks for death, there may be a survivorship bias, and reduced eGFR in the elderly may be more a surrogate for high comorbid disease burden.

Though increased age decreases the likelihood of progression to ESRD and RRT, other risk factors have garnered interest as positive predictors of renal decline. These include male sex, calcium, phosphorus, serum albumin, hemoglobin, blood pressure and others. Of these, baseline eGFR and proteinuria are the most important. Proteinuria is the single most important risk factor for ESRD progression. This has been shown in multiple prospective studies in patients with and without diabetes.

At minimum, one should consider the patient’s age, level of renal function, rate of decline in eGFR, and level of proteinuria when considering who to prepare for RRT.

Timing of renal replacement therapy initiation

Adequate preparation for RRT takes time and the timing of RRT initiation also affects the preparatory period. Patients initiated at higher levels of GFR may have decreased or increased opportunity to prepare for dialysis. However, while recent, studies have shown a trend in initiating therapy at higher levels of GFR these have not demonstrated evidence for improved outcomes although lead time bias may confound these data. Current clinical guidelines vary, however, as to the exact timing of hemodialysis initiation, ranging from an eGFR less than 15 ml/min/1.73 m2 with an allowance for starting earlier based on uremic symptomatology and other comorbities, to a GFR of 6 ml/min/1.73 m2.

Recent trends show that patients are starting at higher levels of renal function. USRDS data from 2008 shows 52% of patients initiating dialysis with an eGFR greater than 10 ml/min/1.73 m2 compared to 20% in 1996 and more patients are starting with an eGFR greater than 15 ml/min/1.73 m2.

This is a trend seen internationally. An international survey of nephrologists in 2000 showed that the principal determinants for starting therapy are uremic signs and symptoms (38%) and residual renal function (32%). Ninety percent of responders felt that starting dialysis 6-12 months earlier would offer some outcome advantages. This change in practice pattern stemmed from earlier studies suggesting harm with late dialysis initiation and published clinical guidelines which sanction RRT at higher levels of GFR.

More recent observational studies and retrospective cohorts have not shown benefit (and may show harm) with early dialysis. Wright retrospectively reviewed outcomes of early or late start to dialysis from the 1995-2006 USRDS database. The early subgroup (eGFR > 15 ml/min/1.73 m2) had greater mortality than the subgroup with eGFR 5-10 ml/min,/1.73 m2, and those starting with eGFR less than 5 ml/min/1.73 m2 least mortality. Comparable results were noted in a relatively healthy cohort of incident nondiabetic dialysis patients ages 20-64 years. Retrospective studies in Canada, Taiwan, and Sweden have reported similar findings as well.

The IDEAL Study (Initiating Dialysis Early and Late) is the first study to randomize patients to early (eGFR 10-14 ml/min/1.73 m2) or late (5-7 ml/min/1.73 m2) start to dialysis. Early initiation of RRT had no affect on rate of death from any cause, cardiovascular events and infectious complications over 3.6 years of follow-up (75.9% of those in the late initiation arm started dialysis above the eGFR target due to uremic symptoms).

Subsequent evaluation of the IDEAL patients shows a trend to higher cost in the early start group without a measurable quality of life benefit. Taken together, these data suggest that patients who are in need of RRT may be able to delay therapy for up to 6 months longer than current practice. This time may be used to optimize readiness for dialysis.

What tests to perform?

Determining kidney function

Factors affecting creatinine generation in the general population are varied based on age, gender and race. Equations estimating GFR and taking these factors into account are now considered preferable for defining renal function and reporting these data is now routine with most laboratories in the United States.

Serum creatinine and Cystatin C

As an amino acid degradation product, creatinine generation is closely linked to muscle mass and dietary factors. It is freely filtered at the glomerulus but also secreted by proximal tubular cells, which renders the creatinine clearance an overestimate of GFR. In addition, some medications may interfere with proximal tubular creatinine secretion (cimetidine, trimethoprim, fibrates) raising serum creatinine (Scr) levels without changing GFR. Specific clinical variables may alter creatinine generation as well including race, amputation, liver disease, muscle wasting, and others.

Cystatin C has received much attention recently as a novel marker for both acute kidney injury and chronic kidney disease, but it also has many influences on its production. KDIGO recommends using Cystatin C as a confirmatory method when serum creatinine is felt to be less accurate.

Estimating equations

Widely accepted estimating equations for GFR in adults include Cockcroft-Gault, MDRD, and more recently CKD-EPI (CKD Epidemiology Collaboration) equation.

The Cockcroft-Gault equation is an estimate of creatinine clearance (CrCl) which uses age, sex, weight in kilograms, and Scr as variables. The equation is CrCl = [(140 – age) x weight)]/(72 x Scr) multiplied by 0.85 if the patient is female. The resulting value is not adjusted for body surface area. Since Cockroft-Gault estimates CrCl, it overestimates GFR.

The MDRD equations are commonly used in clinical practice and regularly reported by reference laboratories. It estimates GFR with adjustment for body surface area. The four-variable MDRD equation varies with age, Scr, sex, and whether the patient is black. Generally, the MDRD equation is considered as accurate or more accurate than either serum creatinine measurements or the Cockcroft-Gault equation.

There are, however, certain circumstances that favor the use of creatinine and urea clearance measurements over estimating equations. Patients with unusual dietary habits or abnormal body size should be considered for direct measurements. In particular, patients with muscle wasting may have less GFR than reported by estimating equations. Unusually muscular patients will have a reported eGFR that underestimates true renal function.

CKD-EPI is a more recently developed equation for estimating GFR and has been recommended to replace MDRD. The equation was developed with data from10 studies and validated with 16 other studies. It appears to have greater accuracy at estimating GFR at more normal renal functions limiting erroneous diagnoses of CKD, and it seems to have greater predictive power in patients with earlier levels of renal dysfunction. The function of MDRD and CKD-EPI appear comparable at lower levels of kidney function (less than a GFR of 45 ml/min 1.73 m2).

Prediction tools

Given that rate of change in GFR over time and proteinuria are predictive of transitioning to ESRD, these values should be measured and followed over time. Usually, renal function can be assessed with MDRD or CKD-EPI and proteinuria can be quantified with a spot urine albumin to creatinine ratio (ACR). This should suffice for risk assessment.

Recent ESRD predictive tools use readily accessible laboratory values including serum phosphate, calcium, bicarbonate, and albumin. Simultaneously, clinicians evaluate and treat conditions associated with advancing renal failure including proteinuria, anemia, metabolic acidosis, electrolyte abnormalities, lipid abnormalities, and hyperparathyroidism with vitamin D deficiency. Treatment of these disorders is addressed elsewhere.

Uremic symptoms

The timing of appearance of overt signs and symptoms of uremia is variable. Some patients remain relatively asymptomatic despite having a GFR below 10 ml/min/1.73 m2 while others become uremic at higher levels.

Common symptoms of uremia include loss of appetite with morning nausea and loss of taste as well subtle cognitive decline and altered sleep cycles. Patients may have volume derangements (hyper- or hypovolemia) and may develop electrolyte abnormalities refractory to medical management.

Indicators for the prompt initiation of dialysis include pericarditis, refractory hypertension or volume overload, refractory hyperkalemia or acidosis, advanced uremic encephalopathy or neuropathy, severe bleeding diathesis, or severe nausea and vomiting.

Monitoring for malnutrition

Malnutrition and hypoalbuminemia at the onset of RRT is associated with increased mortality. Guidelines recommend routine monitoring for protein malnutrition with the prompt initiation of RRT if it cannot be ameliorated with dietary interventions or other correctable factors.

Laboratory, clinical and symptom assessment should increase in frequency as renal function declines. Patients with a GFR less than 15 ml/min 1.73 m2 should be monitored at least every 1 to 2 months.

How should patients in need of renal replacement therapy preparation be managed?

Education and preparation for RRT is a challenging endeavor for patients, families, other caregivers, and treating clinicians. Patients face difficult decisions about goals of care and modes of treatment for ESRD. Confronted with declining health, patients must decide on aggressive therapies (such as dialysis and transplantation) or conservative therapy and end of life care options.

If renal replacement therapy is chosen, patients and clinicians will need to consider type, location, and access for RRT. This section discusses the management for patients considering renal replacement therapy. Specifically addressed are recommendations involving shared decision-making, pre-dialysis education, options for the treatment of advanced renal failure, and dialysis access considerations.

Shared decision making

The Renal Physicians Association has published clinical practice guidelines focused on shared decision-making in dialysis initiation and withdrawal. Establishing this type of relationship with patients is paramount. The benefit of shared decision-making model is that it fulfills the “ethical need to fully inform patients about the risks and benefits of treatments, as well the need to ensure that patients’ values and preferences play a prominent role.” Essential participants include the patient, physician, and family members or caregivers (with the patient’s consent).

Patients should name a surrogate decision maker (to be included in treatment discussions) who can act on behalf of the patient if they lack decision-making capacity. Furthermore, patients and their caregiver team should be fully informed about their diagnosis, prognosis, and all available treatment options.

Educating patients and caregivers is a lengthy, time intensive process that has tremendous impact on treatment plans.

Pre-dialysis education

An integral component of shared decision-making is education about treatment options. Though critical, the delivery of renal replacement education is often suboptimal. In one study, nearly 36% of patients were unaware of kidney disease and nephrology four months prior to initiating therapy. Moreover, education about treatment options occurred within 1 month of dialysis or after dialysis initiation in 48% of patients.

Currently the format of pre-dialysis education varies by location. Standard components include education about renal failure and prognosis, identification and prevention of uremia, and preservation of residual renal function. ESRD treatment options may include conservative medical management with end of life care, transplantation, in-center (including nocturnal if available) or home hemodialysis, and peritoneal dialysis.

A 1993 National Institutes of Health consensus conference suggested that this educational and preparatory process may be best performed in a multidisciplinary setting that often includes nephrologists, nurse educators, dietitians, social workers, and various other providers. This approach is recommended in all clinical guidelines. Recently enacted, the 2008 Medicare Improvement for Patients and Providers Act (MIPPA) promotes pre-dialysis education by paying clinicians for up to six sessions on kidney disease education.

Pre-dialysis education significantly affects treatment modality selection and patient quality of life. It also reduces costs associated with initiating therapy, unplanned dialysis starts, and hospital days in the first month of RRT. Moreover, mortality is likely to be reduced.

Patients will have interest in differing treatment modalities and the impact on mortality and quality of life. Clinicians and patients may also struggle with the choice of RRT or palliative care.


For acceptable candidates, transplantation offers a significant survival advantage over other RRT modalities. In addition, patients are more likely to function at normal levels (79% for transplants compared to 47.5% for hemodialysis and 59% for peritoneal dialysis), are nearly three times as likely to work, and have consistently higher measures of life satisfaction and well-being.

Transplantation is the optimal choice for patients nearing ESRD. If performed pre-emptively, dialysis associated morbidity is avoided, healthcare costs are reduced, and graft and patient survival are increased. A further survival advantage exists for living donor transplantation compared to deceased donor transplantation. It has also been observed that transplantation greater than 180 days after the initiation of dialysis is associated with increased graft loss.

Despite these data, pre-emptive transplant remains an infrequently used treatment choice. The USRDS report noted that 60% of incident ESRD patients were followed by nephrology at least 6 months before initiating RRT, but only 5.7% of them were wait-listed for a transplant at the time of initiation of RRT and only another 0.8% received a living donor transplant without being listed. Overall, less than 3% of patients with ESRD undergo transplantation as the initial RRT modality. Weng and colleagues have observed that many patients have the time and opportunity to be evaluated for pre-emptive transplantation but it is frequently not actualized.

To improve rates of pre-emptive transplant referral, guidelines recommend simultaneous referral to transplantation in appropriate patients at the time of vascular access referral. Patients should receive early education about transplantation with a goal of having them wait-listed when GFR falls below 20 ml/min/1.73 m2, and special attention should given to identifying and evaluating potential living donor candidates.

Peritoneal dialysis and hemodialysis

As very few patients currently receive pre-emptive transplantation, most patients will start peritoneal dialysis (PD) or hemodialysis (HD). Each modality has distinct advantages and disadvantages.

Peritoneal dialysis offers patients a home modality that is portable, requires fewer clinic visits, and allows patients to actively manage their ESRD. It furthermore has comparatively minimal affects on patient hemodynamics acutely, which reduces recovery time after dialysis. Those with needle aversion will also prefer this technique. PD is a continuous modality performed daily (either extended periods overnight or multiple times during the day) necessitating rigorous patient and caregiver training and motivation. A sanitary home environment with ample storage capacity (for medical supplies) is also critical.

In-center hemodialysis can be provided to patients who do not wish to perform self care or do not have an appropriate home environment. In-center therapy also allows sicker patients to be closely monitored. In center HD is associated with hemodynamic changes, volume and solute shifts which contribute to post-treatment malaise. It also requires coordinated travel to facilities and is subject to inflexible scheduling.

Home hemodialysis therapy is an option for patients which combines the independence of home care and self management of hemodialysis.

The prevalence of each modality varies by location internationally. PD accounts for close to 80% in Hong Kong and Mexico but only a small percentage in the US and in many European communities. The application of PD and HD modalities in the elderly varies substantially by location as well.

The most important factors to modality selection include medical or social barriers to home dialysis and patient education about different options. As expected, those educated about home dialysis therapies are much more likely to choose one of these options.

Quality of life (QOL) comparisons do not favor one modality over another.

Prospective cohort studies are limited and with relatively small numbers of patients. NECOSAD was a prospective cohort study comparing PD and HD in incident dialysis patients from the Netherlands showing no difference in the relative risk of death at 2 years with a slight advantage to HD after 2 years. United Kingdom and Canadian prospective cohort studies also failed to show outcome differences between HD and PD.

The mortality and QOL differences between PD and HD are small. HD may be slightly more favorable in elderly patients and diabetics, while PD may be more favorable in young nondiabetics in terms of mortality and morbidity outcomes. Personal preference, home environment, physical and mental health, and modality education are other important factors in deciding between PD and HD.

Conservative and palliative care

Appropriate patients should be fully informed about the option for palliative care along with other modalities. Unfortunately, specific discussions surrounding palliative care and prognosis often neglected. Hindering factors include patient cognition, age, race, perception of prognosis, and timing of nephrology referral.

Perception of prognosis has been shown to vary widely among clinicians. Recent studies however have identified patient populations who have particularly poor outcomes with dialysis. Informing these patients of their prognosis may assist decision making.

The elderly represent a growing segment of the ESRD population with the incidence of treated ESRD being the highest among patients at least 70 years of age. This has increased substantially over the last several decades (57% from the period of 1996-2003 by USRDS data). Multiple factors contribute to this trend, including rising prevalence of CKD and diabetes in the population, improved cardiovascular care, earlier initiation of dialysis, and the rise of incident AKI.

Elderly patients have greater levels of comorbidity and reduced life expectancy on dialysis. Recently, outcomes in this population have received attention. In 2009, Tamura and colleagues reported outcomes in nursing home patients who initiate dialysis. USRDS registry data was cross-referenced against national nursing home registry data to identify and assess the functional status of this subset of patients before and after initiation of dialysis.

The mean age of the population was 73 years, 69% initiated dialysis while hospitalized at an acute care facility. At 3 months, only 39% of patients had maintained their pre-dialysis functional status (as measured by MDS-ADL, Minimum Data Set – Activities of Daily Living), and by 12 months, 58% of the patients had died with only 13% maintaining their pre-dialysis functional status.

Beyond the elderly nursing home resident, multiple studies demonstrate poor outcomes in the elderly population in general. USRDS registry data has shown that patients 80 years and older have 46% 1-year mortality and a 20% 3-month mortality. These high mortality rates are confirmed in other elderly cohorts. In some studies, the single most common cause of death was withdrawal from dialysis (38%), which has become the second leading cause of death in the elderly dialysis population.

Factors such as decreased ambulatory status, peripheral vascular disease, and ischemic heart disease have a significant negative impact on patient outcomes. The RPA guidelines note that patients with any two of the following characteristics have a particularly poor outcome: age greater than 75 years, high comorbidity score (a modified Charlston Comorbidity Index score of 8 or greater), marked functional impairment (a Karnofsky Performance Status Scale score of less than 40) or severe chronic malnutrition with serum albumin level less than 2.5 g/dl.

In 2010, Davison described essential components of end-of-life care practices for patients with CKD. Patients should be educated about their prognosis and renal replacement options and assessed for palliative care needs (those choosing conservative management or having a high 1-year mortality risk). They should establish an advanced care plan. Additionally, screening and management of pain and other symptoms should be performed while also considering patient needs for spiritual, emotional and psychosocial support. Access to palliative care treatment options should be available.

Hemodialysis access

For patients with progressive renal failure who plan to have hemodialysis, an important component of preparation is access planning and creation. Multiple studies have shown that appropriate access at initiation of dialysis improves survival and minimizes complications.

Conversely, the use of catheters is associated with excess morbidity, mortality, and cost. There is an increased risk of death within the first 3 months of RRT in patients using central venous catheters (CVC). Additionally, those switching from CVC to arteriovenous (AV) fistula use during the first three months had intermediate risk of death.

Guidelines recommend AV fistulas as the preferred access for HD. Distal radiocephalic wrist fistulas are initially favored over upper arm brachiocephalic and brachiobasilic vein fistulas. Distal arm fistulas are the preferred choice because they preserve proximal fistula options for future consideration.

Even with optimal preoperative assessment and vein protection, some patients (elderly, diabetic, and female) have low radiocephalic AV fistula success. They should be considered for upper arm brachiocephalic or brachiobasilic fistula creation.

Once AV fistula options have been exhausted, AV graft placement should be considered. Central venous catheters should be avoided unless other options are unavailable.

Arteriovenous fistulas are preferred over other access options because they have lower thrombosis and intervention rates and have lower rates of infection. Moreover AV fistula use is associated with substantial increased survival (RR of death for CVC is 2.3, and AV graft is 1.47), lower rates of hospitalization, and lower cost.

Through the Centers for Medicare and Medicaid Services (CMS) Fistula First Breakthrough Initiative (FFBI), the prevalent use of AV fistulas has significantly increased. The new FFBI target for prevalent AV fistula use is 66%. This target is comparable to achieved rates in several European countries however there is debate about whether this is a realistic target in the U.S. As of October 2010, the rate of AV fistula use among hemodialysis patients in the U.S. was 57.1%, with one renal network having achieved the target of 66%.

There is variance in AV fistula use across countries and within countries. Factors influencing the rate of AV fistula creation and use may include timing of nephrology referral, variability in the presence of a dedicated vascular access coordinator, patient education about the benefits of AV fistulas, lack of insurance, differences in dialysis prescriptions (lower blood flow rates are more common in Europe), and dialysis staff cannulation training.

Though prevalent rates of AV fistula use are on the rise, incident hemodialysis fistula use remains low. The 2008 USRDS data show that only 30% of incident dialysis patients used an AV fistula on their first outpatient dialysis run and another 16% had a maturing AVF. Additionally, pre-ESRD fistula placement is low (15%). Over 80% of patients initiate dialysis with a CVC.

Surgical access evaluation

There is debate about when patients should be referred to surgery. Current clinical guidelines, which are largely opinion based, recommend referral for fistula placement when serum creatinine is over 4 mg/dl, GFR is less than 25 ml/min/1.73 m2, or time to dialysis is estimated to be less than one year, or referral at least 6 months before anticipated need for dialysis.

Preoperative assessment for access creation has become more complex. It is now standard of care to perform arterial and venous mapping by ultrasonography to optimize adequate AV access creation. Mapping has substantially increased the success rate of AV fistula placement.

In distinction to central venous catheters, which can be used immediately, and AV grafts which can often be used within weeks, creation and maturation of an AV fistula may take 2 to 4 months or longer. Furthermore, up to 20-50% of AV fistulas fail to mature) or require at least one or two percutaneous interventions to ensure successful use.

To be considered adequate, an access should have a minimum blood flow rate of 350-450 ml/min which can be maintained throughout a 3- to 5-hour dialysis session. Guidelines recommend that an AV fistula may be considered mature enough for use when it meets the rule of 6’s (blood flow greater than 600 ml/min, diameter greater than 0.6 cm, and depth no more than 0.6 cm). Many non-maturing fistulas can be identified by clinical exam or imaging techniques within 4 to 6 weeks.

The most common findings in a non-maturing access include access stenosis (limiting blood flow through a portion or all of the AV fistula) and accessory veins (stealing blood flow from the dominant vein). Hence, clinicians need to refer patients with enough time to undergo preoperative mapping, surgical consultation and surgery, postoperative maturation assessment, and intervention if a non-maturation is identified.

PD catheter evaluation

For peritoneal catheter insertion currently the standard surgical technique uses a silicone rubber based swan neck catheter with dual polyester velour cuffs inserted through the rectus sheath rather than midline. Moncrief and colleagues have described a technique for burying the external portion of the catheter at the time of catheter insertion.

Timely and skilled insertion remains a barrier to more widespread PD use in many centers. A recent survey of surgical training centers showed that most surgical residents perform less than five peritoneal catheter insertions during their training and that training programs do not emphasize this procedure. Local surgical expertise is perhaps the best way to guide the type of catheter and the surgical technique used.

The appropriate timing for PD catheter insertion is unclear. Due to shorter maturation times, most patients will have PD catheters inserted much closer to the time of planned PD initiation. Using PD catheters too soon after insertion is associated with complications such as leakage and infection. On the other hand, delaying PD catheter insertion and use exposes patients potentially to the morbidity and expense of HD. Most clinical guidelines suggest catheter use no sooner than two weeks after catheter insertion.

Recent nonrandomized studies show that early use of PD catheter with low PD volumes and limited patient activity may not significantly increase the risk of complications.

Patients should have early evaluation by surgery to assess the possibility of PD catheter placement with a plan for surgical catheter insertion 4 to 6 weeks prior to the anticipated initiation of PD. The Moncrief technique, however, allows for catheter placement up to 6 months before the anticipated date of PD initiation.

How to utilize team care?

Educating patients about renal replacement options and treatment goals in preparation for RRT is a multidisciplinary venture orchestrated by the nephrologist. In 1993, a National Institute of Health consensus conference proposed that patients with CKD be referred to a renal team consisting of a nephrologist, dietitian, nurse, social worker and a mental health professional. Recently in the U.S., the Medicare Improvements for Patient and Providers Act (MIPPA) of 2008 has provided limited funding for improved patient pre-dialysis education.

The evidence showing benefit of multidisciplinary clinic education and preparation is increasing. Retrospective Canadian studies also show significant benefits to multidisciplinary pre-dialysis education. These studies showed fewer unplanned starts, greater success with early access creation, less inpatient days at the start of dialysis, and significant overall cost savings. Similar results have been noted with physician-extender run CKD clinics.

A multidisciplinary clinic focused on advanced CKD care can provide the infrastructure for managing the complex care and education required. Not only can patients be assessed for therapeutic goals (optimized blood pressure management, anemia evaluation and treatment, and treatment of metabolic bone disease and acidosis for example), they can be educated about RRT options (including conservative care), access needs, and complex dietary treatment plans.

Many centers have established a dedicated nursing vascular access coordinator in attempt to improve rates of AV fistula use. As a quality improvement measure, Polkinghorne and colleagues showed improvement in AV fistula use from 56% to 75% with a decline in the number of catheter days by 50% after creation of a dedicated vascular access nurse coordinator. Others have shown that multidisciplinary access meetings between surgery, nephrology, and interventional radiology have improved primary and secondary patency rates.

Specialty consultations

Timing of vascular access referral

As previously detailed, timely access consideration is essential for those patients considering hemodialysis. Once settled on hemodialysis, patients must be referred to surgery and undergo preoperative arterial and venous sonographic mapping. After access creation, it must be followed within 4-6 weeks for maturation and intervened upon quickly if failure of maturation is found. Minimizing delays in this pathway is important to maximize incident HD AV fistula use. Empowering a vascular access coordinator to minimize delays will increase the likelihood that patients will initiate HD with a functional AV access.


Patients who are potential candidates for renal transplantation should be referred at least when the access referral is being made. There are many benefits to pre-emptive transplantation, including improved overall and graft survival and improved quality of life. Unfortunately, very few patients receive pre-emptive transplantation.

Diet and nutrition

Dietary counseling can have a significant impact on care in advanced stages of CKD. Patients need tight dietary restriction of sodium and fluid to minimize symptoms of volume overload and hypertension, dietary potassium restriction to limit hyperkalemia, and dietary protein restriction to limit uremic symptoms and possibly delay the onset of RRT. As might be expected, patients also need frequent assessment for malnutrition.

Expert opinion varies on the use of dietary protein restriction for patients with advanced CKD. Proponents suggest that it may be performed safely to reduce rate of eGFR decline, limit uremic symptoms, and delay initiation of RRT. It also does not appear to increase the risk of malnutrition and may paradoxically limit the malnutrition due to careful dietary monitoring. Compliance however is expected to be less than 50% and opponents suggest favorable results may be more a result of publication bias.

Some recommend that for a eGFR less than 25 ml/min/1.73 m2, patients should be prescribed a diet containing 0.6 g/kg/day of protein and nutritional status should be evaluated by a variety of measures every one to three months. Patients not yet on dialysis should have caloric intake of 35 KCAL/kg/day if less than 60 years and 30 kcal/kg/day if greater than 60 years.

It is recommended that patients have albumin, body weight assessment, and/or subjective global assessment (SGA) measured every 1-3 months. Dietary interviews and diaries should be performed every 3-4 months with increased monitoring as renal failure worsens. Dietary restriction and patient monitoring will require dedicated dietary expertise and motivated patients. This is best achieved through a multidisciplinary clinic referral to dietary experts within the community.

Are there clinical practice guidelines to inform decision making?

  • Caring for Australians with Renal Impairment, CARI-Australia

  • National Kidney Foundation K/DOQI Guidelines

  • Canadian Society of Nephrology – CSN

  • European Renal Association – European Dialysis and Transplant Association (ERA-EDTA) European Best Practice Guidelines (EBPA)

  • UK-Renal Association

  • Kidney Disease Improving Global Outcomes (KDIGO)


  • Renal Physician Association: Shared Decision-Making in the Appropriate Initiation of and Withdrawal from Dialysis Clinical Practice Guideline, Second Edition.

Other considerations