2009 USRDS Annual Data Report
This Chapter
Download PDF: WIN*   MAC*
Download all slides: WIN*   MAC*
*corresponding data in Excel included
Search This Page
Search All

Three - Patient characteristics

Using data from the Medical Evidence form (2728), the USRDS tracks patient information at the initiation of end-stage renal disease (ESRD) treatment, assessing changing demographic characteristics and access to care. A revised form was introduced in 2005, with new information on vascular access use at the first outpatient dialysis treatment as well as on care received prior to ESRD from nephrologists and nutritional counselors. The high first-year mortality rates reported in recent ADRs have raised questions about risk factors that may contribute to elevated morbidity and mortality. Last year we began reporting on the vascular access used at the first outpatient dialysis, noting the high percentage of patients using a catheter. New data for 2007 show that more than 80 percent of patients use a catheter at initiation; 20 percent have a catheter along with a maturing internal vascular access. Findings differ, however, based on primary diagnosis. In patients whose ESRD is caused by some form of cystic kidney disease, the rate of catheter use at initiation is just 61 percent. This same population also has the lowest percentage of patients receiving no nephrologist care prior to ESRD, and the greatest percentage receiving such care for more than a year. These data suggest that early nephrologist care is associated with better planning for vascular access placement, leading to lower use of catheters and, potentially, to lower infectious complication rates as well. The level of anemia correction after the initiation of dialysis has slowed, with the rise in hemoglobin levels now similar to that seen in 2004 — a notable change from the increasing levels presented in past ADRs. There also appears to be a shift in the pre-ESRD use of erythropoiesis stimulating agents (ESAs), with a continued decline since 2004. This shift may, however, simply reflect a change in the ESA question on the new Medical Evidence form. The previous form asked if an ESA was used before dialysis began, which could be misinterpreted to mean before a single dialysis treatment rather than before the ESRD first service date, as intended. The form now asks specifically if ESA therapy was begun before ESRD. Across all age and race groups, creatinine levels at the initiation of ESRD therapy continue to decline, and are lowest in patients age 75 and older. There has been concern that the lower serum creatinines at initiation may indicate that physicians are choosing to start therapy earlier in the course of disease. Other possible explanations could center on the increasing survival of CKD patients, with more now considered for dialysis than in prior years. The large increase in the number of patients from the baby boomer generation (age 45–64) appears to be driving the continued growth of the ESRD population, with declining rates of cardiovascular mortality meaning that more patients survive to reach nephrologist care and ESRD treatment. These possible causes underlying the initiation of ESRD patients with lower serum creatinines and higher estimated glomerular filtration rates (eGFRs) will require additional investigation to determine if the changes are related to physician practices or to changes in CKD patient survival, leading to greater early referral to nephrologists. The recently revised Medical Evidence form includes new questions on laboratory data prior to ESRD initiation. Among patients beginning ESRD therapy in 2007, the average cholesterol and triglyceride levels were 157 mg/dl and 154 mg/dl, respectively, and levels were highest in children. Average glycemic control appears to be at a glycosylated hemoglobin (A1c) level of 7.3 percent. This suggests that one in two diabetic patients has less than optimal glycemic control, based on guidelines from the American Diabetes Association which target an A1c level of less than 7 percent. Patients with a primary diagnosis of glomerulonephritis or cystic kidney disease, and those with diabetes as a secondary comorbidity, have an average A1c of 6.3–7.1 percent. This may reflect severity of disease, as those with secondary diabetes may have had the disease for a shorter period of time and therefore have fewer complications. In conclusion, the incident population age 45–64 is expanding. The population appears to carry more comorbidity, and is starting ESRD therapy with higher eGFRs — a reflection of both increasing disease burden and, in those with low body mass indices, of malnutrition as well. The use of anemia treatment before dialysis appears to have declined, but some of these changes may reflect a change in wording on the newest Medical Evidence form. Correction of anemia after the initiation of dialysis is now similar to that noted in 2004, a finding that may reflect practice changes secondary to safety concerns regarding higher hemoglobin levels noted in recent clinical trails, changes in FDA labeling for ESAs, and CMS payment policies for hemoglobin levels over 13 g/dl. Lipid abnormalities are present at the start of dialysis, and are more pronounced in younger patients. And glycemic control appears to be an issue in the population starting dialysis with diabetes as a primary or complicating condition. Younger patients, who typically have Type 1 diabetes, have less control than older patients, who typically have Type 2.

Figure 3.1; see page 366 for analytical methods. Incident hemodialysis patients, 2007, with a valid Medical Evidence form; access type & primary diagnosis obtained from the Medical Evidence form.

The number of new ESRD patients who receive no care from a nephrologist prior to beginning therapy varies by primary diagnosis, from just 18.5 percent of those with cystic kidney disease to more than 46 percent of those whose ESRD is caused by hypertension. Figure 3.2; see page 366 for analytical methods. Incident ESRD patients, 2007.

Among hemodialysis patients starting ESRD therapy in 2007, 57 percent of those with a catheter had received no nephrologist care prior to initiation, compared to 10 percent of those with an arteriovenous fistula, and 21 percent of those with an arteriovenous graft. Figure 3.3; see page 366 for analytical methods. Incident hemodialysis patients, 2007.

This figure looks at new patients who received more than 12 months of nephrologist care prior to starting ESRD therapy in 2007. Fourteen percent of those with a primary diagnosis of diabetes or hypertension began treatment with an arteriovenous fistula, compared to 33 percent of those with ESRD caused by cystic kidney disease. Figure 3.4; see page 366 for analytical methods. Incident hemodialysis patients, 2007.

Among patients beginning ESRD therapy in 2007, nearly 43 percent had not seen a nephrologist prior to initiation. Of those with no pre-ESRD nephrologist care, 91 percent had a catheter as their inital access, while only 2.9 and 1.7 percent had a fistula or graft. Eleven percent had a catheter with a maturing fistula, and 2.3 percent had a maturing graft. Twenty-four percent of patients with more than twelve months of nephrologist care, in contrast, used a fistula at initiation. Use of erythropoiesis stimulating agents (ESAs) and dietary care are more common in patients with nephrology care. In those seeing a nephrologist for more than one year, for example, 54.4 percent were on an ESA pre-ESRD and 18 percent saw a dietitian, compared to 3.4 and 0.2 percent, respectively, of those with no pre-ESRD nephrologist care. Table 3.a; see page 366 for analytical methods. Incident ESRD patients, 2007.

In 2007, hemoglobin levels following initiation in incident dialysis patients were unchanged from those found in 2004. Levels increased from 10.5 to 12.4 g/dl over the first four months of dialysis — 1.2 and 0.4 g/dl higher, respectively, than levels achieved in 1998 and 2001, and slightly higher than the recommended target hemoglobin level of 11–12 g/dl. In 2004 and 2007, the rapid rise in hemoglobin levels following initiation indicates a willingness on the part of renal practitioners to aggressively manage anemia early in the dialytic process. With recent clinical trials (CHOIR and CREATE) showing evidence of potential harm in high hemoglobin levels, and KDOQI guidelines indicating no benefit related to morbidity or mortality with a hemoglobin above 13 g/dl, anemia management protocols must incorporate safeguards against overuse of erythropoiesis stimulating agents (ESAs) and IV iron so that that patients do not remain at high hemoglobin levels for an extended period of time.

In males and females, 2007 hemoglobin levels four months post-initiation were 1.1 and 1.2 g/dl higher, respectively, than 1998 levels, while levels in whites, African Americans, and patients of other races rose 1.1, 1.3, and 1.1 g/dl, respectively.

In the incident ESRD population, the mean hemoglobin level at initiation of therapy peaked in 2006 at 10.2 g/dl, and has since fallen to approximately 10.0 in May, 2008 — meaning that half of new patients begin therapy with a hemoglobin lower than 10 g/dl. This coincides with a decline in the proportion of patients receiving pre-ESRD ESAs, from a high of 33–34 percent in 2002–2004 to 26–27 percent in the first part of 2008.

The percentage of patients receiving ESAs prior to initiation is highest for those residing in the northern tier of states, the New England area, and portions of the southwest, the Ohio Valley, and Texas, averaging 43.1 percent in the upper quintile. Figures 3.5. 3.6, & 3.7; see page 366 for analytical methods. Incident dialysis patients (3.5); incident ESRD patients (3.6–7).

Cholesterol values at the start of ESRD therapy vary by race, from a mean of 147 mg/dl in Native American patients to 167 among African Americans. The range is even greater by age, from 144 mg/dl in the oldest patients to 192 in the pediatric population. Initial glycosylated hemoglobin (A1c) values suggest that blood glucose levels are best controlled in the Native American population and in patients with glomerulonephritis and cystic kidney disease. Table 3.b; see page 366 for analytical methods. Incident ESRD patients, 2007. *A1c data include only patients with diabetes as their primary diagnosis or as a comorbidity.

Serum creatinine levels at the start of ESRD treatment have fallen 2.4 mg/dl since 1995, from 8.7 to 6.3, perhaps indicating a tendency to initiate ESRD therapy earlier in the course of a patient’s treatment, or suggesting that patients are starting treatment with greater disease burdens. In 2007, levels ranged from 5.2 mg/dl among patients age 75 and older to 8.4 in those age 20–44. By race and ethnicity, the initial mean creatinine level ranges from 5.8 mg/dl among whites to 7.4 among African Americans.

One in five patients beginning ESRD therapy in 2007 did so with a serum creatinine of less than 4 mg/dl, while nearly six in ten had a level of 4.0–7.9. The percentage of patients initiating with a level less than 4 mg/dl rises by age in adults, from 8.8 in those age 20–44 to 29.4 in those age 75 and older, and is 25.8 in women compared to 14.6 in men. Figures 3.8, 3.9, 3.10; see page 366 for analytical methods. Incident ESRD patients.

The percentage of new adult patients with an estimated glomerular filtration rate less than 5 ml/min/1.73 m2 declines with age, from 16.6 among those age 20–44 to just 4.5 among the oldest patients; this coincides with the rise in serum creatinine levels by age noted in Figure 3.10. Nearly 18 percent of patients initiate with an eGFR of 25 or greater. Figure 3.11; see page 366 for analytical methods. Incident ESRD patients, 2007.

Patients initiating ESRD therapy continue to begin with less than optimal hemoglobin levels. In 2007, one-half of new patients had a hemoglobin less than 10 g/dl; only 12.5 percent began therapy with a hemoglobin of 12 or above. This in part reflects the low number of patients receiving an erythropoiesis stimulating agent prior to initiation — just 27 percent, as shown in Figure 3.6. Figure 3.12; see page 366 for analytical methods. Incident ESRD patients, 2007.

Glycosylated hemoglobin (A1c) levels above 7 percent indicate poorly controlled diabetes. Nearly four in ten diabetic patients starting ESRD therapy in 2007 had an A1c level of 7 percent or above. This varies little by gender, but by age ranges from 27 percent of those age 75 and older to 55 percent of those age 20–44. Figure 3.13; see page 366 for analytical methods. Incident ESRD patients with diabetes as their primary diagnosis or as a comorbidity, 2007.