Management of Patients with Chronic Renal Insufficiency in the Northeastern United States

The prevalence of end-stage renal disease (ESRD) in the United States increased by 7% per year between 1992 and 1996 and was 306,967 at the end of 1997 (1). The morbidity and mortality among patients with ESRD is high, and the care of these patients consumes a significant proportion of health care resources (2,3). Clinical and economic outcomes among patients with ESRD are influenced by a number factors, such as age, gender, race, malnutrition, and nonrenal comorbid conditions (4,5,6). However, comorbid conditions begin to develop early during the course of chronic renal insufficiency (CRI). It was estimated recently that 2.5 million individuals in the United States have serum creatinine levels ≥1.7 mg/dl and 0.8 million have serum creatinine levels ≥2.0 mg/dl (7). Timely and optimal care during CRI could modify the severity of comorbid conditions and the high mortality among patients with ESRD. Furthermore, improved and timely care during CRI possibly may improve the quality of life of patients with renal insufficiency, reduce the incidence of ESRD, and reduce the cost of ESRD care (8,9). Thus, optimization of care during CRI could be the key to improved ESRD outcomes.

Although formal guidelines for care of the patient with CRI do not exist, few would argue that optimal care should involve early diagnosis of renal insufficiency and use of interventions to delay its progression, prevention or attenuation of the expected clinical and biochemical abnormalities of renal dysfunction, modification of comorbid conditions, adequate preparation for ESRD therapy, and timely initiation of dialysis. However, there is limited information on the patterns of testing of CRI patients for prevalence of complications of CRI and nonrenal comorbidity or the practice of the above-mentioned interventions. Hence, we performed a historical prospective cohort study to investigate the current status of care during CRI in the Greater Boston area.

Materials and Methods

Results

A total of 4016 patients were followed in the nephrology practices during the study period. A total of 682 (17%) met the study criteria for CRI; of these, 602 (88%) had data that could be evaluated. The mean length of follow-up among patients with two or more visits during the study period was 428 d (median 273 d). Selected details of baseline demographic, clinical, and laboratory characteristics are presented in Table 1. Overall, 28% of the patients had diabetes and 25% had HTN as cause of CRI. However, 87% had HTN as a documented condition. The majority of patients had an IDS score of 2 (85%), and only 11% of patients had an IDS score of 3. The majority of the patients had HMO insurance (64%), 13% of patients had private insurance, 17% had Medicare, 4% had Medicaid, and only 2% of patients had no insurance (details not shown).

Timing of First Visit to Nephrology

The first nephrology visit occurred, on average, when the serum creatinine was 3.1 mg/dl and the GFR was 22.2 ml/min per 1.73 m². The distributions of serum creatinine levels and predicted GFR at the first nephrology visit are presented in Figure 1. Renal function at first nephrology visit varied slightly by patient demographic and clinical characteristic, but the differences were not statistically significant (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

(A) Distribution of serum creatinine at the first visit to the nephrologist among patients with chronic renal insufficiency (CRI). (B) Distribution of GFR at the first visit to the nephrologist among patients with CRI. GFR was predicted by use of the equation derived from the Modification of Diet in Renal Disease Study (11).

Discussion

It was estimated recently that 800,000 patients in the United States have serum creatinine levels ≥2 mg/dl and 6.2 million have serum creatinine levels ≥1.5 mg/dl (7). The Modification of Diet in Renal Disease Study demonstrated that renal function declined progressively at an average rate of 4 ml/min per yr in the majority of patients with GFR levels <55 ml/min per 1.73 m² (12). Taking these two studies together and that the U.S. population is aging (13), it can be predicted that the ESRD population will continue to increase in the future. The CRI population already under the care of physicians can be most readily targeted for interventions to improve the outcomes of patients with CRI and ESRD and reduce the costs associated with ESRD care.

The prevalence of CRI in the nephrology practices was 17%. The age and gender distribution of the CRI patients in this study was similar to that of patients starting ESRD therapy in the United States. The prevalence of HTN as the cause of CRI also was similar to the prevalence of HTN as the cause of ESRD in the United States. However, only 28% of the patients had diabetes as the cause of CRI, compared with 44% of patients starting ESRD therapy in the United States in 1997 (1). The mean serum creatinine and predicted GFR at the first nephrology visit were 3.1 mg/dl and 22.2 ml/min per 1.73 m², respectively. To our knowledge, the only published recommendation to date regarding timing of nephrology referral is the NIH 1994 Consensus Statement on Morbidity and Mortality of Dialysis, which suggests referral to a nephrology team when the serum creatinine is >1.5 mg/dl for women and >2.0 mg/dl for men (14). However, the creatinine level at which a nephrologist should first become involved in the care of patients with renal dysfunction is a hotly debated issue. Nevertheless, it is becoming increasingly clear that complications of CRI, such as hypoalbuminemia, abnormalities of calcium-phosphorus metabolism, and cardiovascular disease, begin early in the course of disease and are progressive in nature (15,16,17,18). Indeed, we found that 38% of patients had severe anemia (HCT <30%) during the study, and, among patients who were screened, 55% had or were treated for abnormalities of calcium-phosphorus metabolism, 42% had hypercholesterolemia, 22% had hypoalbuminemia, and 19% had acidosis at some time during the study.

An important aspect of care during CRI is the implementation of interventions to slow the progression of renal disease. ACE-I use has been demonstrated to retard the progression of renal failure among patients with diabetes, and this probably is true among individuals without diabetes as well (19,20). Clinical trials also have demonstrated that BP control, glycemic control among patients with diabetes, and perhaps dietary protein restriction retard the development and progression of renal insufficiency (21,22,23). In the current study, only 49% of patients were found to be on ACE-I at some point during the study follow-up. In particular, among patients with diabetes, only 65% were on ACE-I. This is despite unequivocal evidence that ACE-I slows the progression of CRI (19,24,25) and the National Kidney Foundation's published recommendations for use of ACE-I among patients with diabetes mellitus who have microalbuminuria (26). The low rate of use of ACE-I among patients who are at risk has been reported in other situations. In a study of hospitalized Medicare patients in Atlanta, McClellan et al. (27) found that <35% of patients with diabetes and HTN with renal insufficiency had been prescribed ACE-I at discharge. Sixty-five percent of the patients in the current study were hypertensive, with two or more BP above the target level of 140/90 mmHg recommended in the fifth report of the Joint National Committee on Detection, Evaluation, and Treatment of High BP in 1993 (28). Among patients with diabetes, only 28% had a record of a HbA1C level to monitor adequacy of glycemic control. The results of this study suggest that impressive results from clinical trials on progression of CRI have not been translated fully into clinical practice, even in nephrology practices. However, it is possible that patients in whom a given intervention was not instituted may have had a contraindication for that therapy that was not captured in the data collection. It is also possible that an intervention recommended by the nephrologist may not have been implemented by the referring physician.

Optimal care during CRI includes prevention or attenuation of abnormalities and comorbid conditions associated with renal dysfunction. The majority of patients with progressive CRI invariably develop anemia, malnutrition, and abnormal calcium-phosphorus metabolism, and many have hyperlipidemia, all of which are associated with adverse outcomes among patients with ESRD (18,29,30,31,32,33,34,35,36,37). We found low rates of screening for many comorbid conditions or complications of uremia. Most patients had HCT, serum calcium, phosphorus, and albumin levels recorded, but very few had PTH and serum lipid levels recorded. It must be emphasized that these results may be underestimates, because only testing at the nephrology clinic or after referral to nephrology was captured. Nevertheless, failure to screen suggests lack of identification and treatment of conditions that contribute to increased morbidity among patients with renal failure. In our study, 38% of patients had HCT <30% at some time during the study, but, among these, only 59% received rHuEPO. This compares to a prevalence of HCT <30% of 67% and of rHuEPO use of 23% at initiation of dialysis previously described in the population of patients beginning ESRD therapy in the United States between 1995 and 1997 (38). It is important to note that the Health Care Financing Administration authorizes reimbursement for the use of rHuEPO for patients with renal insufficiency after the HCT falls below 30%, and many other third-party payers follow Health Care Financing Administration policy. Thus, lack of reimbursement for rHuEPO should not be a cause for failure to use rHuEPO when indicated. The prevalence of hypoalbuminemia in the study population was 23%, which compares to a 60% prevalence at initiation of dialysis among patients beginning ESRD therapy in the United States (38). These data demonstrate that the hypoalbuminemia observed among patients starting dialysis begins early during the course of CRI. The prevalence of abnormal calcium-phosphorus metabolism was 55%; among these patients, 60% received a phosphate binder and/or calcitriol. This study confirms that anemia, hypoalbuminemia, and abnormal calcium-phosphorus metabolism are prevalent early during the phase of CRI and that these conditions may require increased attention.

The final component of optimal care during CRI is adequate preparation for ESRD therapy and timely initiation of dialysis. Patients who are educated early regarding the need for RRT are more likely to choose peritoneal dialysis (4,8,39,40). In the current study, only 11% of patients selected peritoneal dialysis, possibly related to the late stage of renal failure at which the first discussion of the need for RRT occurred. However, it must be emphasized that a discussion may have occurred earlier but was not recorded. Timely access placement enables the initiation of dialysis without further procedures in an outpatient or nonemergent manner; thus results in lower use of resources in the initial phases of ESRD (41,42,43). Forty-one percent of the patients who were identified as having progressed to the imminent need for RRT had placement of permanent access before start of RRT, which compares to the rates of >50% for the Michigan (44) and United States (4). population. Last, only 4% of patients who were approaching ESRD were recorded as going on to transplantation, although it is known that patients who have transplants have improved quality of life. Further improvement in patient education, rates of placement of vascular access before the initiation of dialysis, and rates of transplantation are likely to promote a better quality of life for the patients and may result in reduced cost of ESRD care.

There are several limitations of this study, which relate mostly to the source of the patient population. First, the Greater Boston area has a very high penetrance of HMO, which may affect screening and referral practices; thus, the conclusions may not be readily generalized to the U.S. population with CRI. Second, the patient sample was derived from patients who were under the care of a nephrologist. Thus, the management practices of other physicians were not captured, and only the prevalence of CRI and its complications in nephrology practices was estimated, not that of CRI in the general population. Nonetheless, nephrology clinics are where patients with CRI are concentrated and hence are the ideal targets for strategies to improve management. Third, data were collected retrospectively; thus, there is great variability in the information and follow-up available on each patient. Notwithstanding these limitations, this report provides valuable information on a particular population that has not been well studied to date.

In summary, there is a relatively high prevalence of comorbid conditions among patients with CRI, and there seems to be room for improvement in the management of complications of CRI. An argument may be made that early interventions among patients with CRI could result in a slower rate of progression of renal insufficiency, reduction in the severity of renal and nonrenal comorbid conditions, smoother transition to ESRD, and possibly improved patient satisfaction. This ultimately could result in slowing the rate of rise in the incidence of ESRD; decline in the morbidity, mortality, and cost of ESRD care; and a better quality of life for patients. Evidence in support of the above could trigger significant changes in the practices of physicians who care for such patients and encourage third-party payers to commit resources toward early and comprehensive treatment of patients with CRI. The increased costs associated with early nephrology referral and intervention could be offset by reductions in the cost of ESRD care, in particular, the high cost in the first 6 mo of RRT. Future studies will be needed to accumulate evidence in support of the assumptions made above. Meanwhile, the nephrology clinic should be the first target for institution of care pathways to improve the clinical management of patients with CRI.