Nephrol Dial Transplant (2006) 21: 3458 3465 doi:10.1093/ndt/gfl428 Advance Access publication 28 July 2006 Original Article Predictors of worsening renal function in adult patients with congestive heart failure receiving recombinant human B-type brain natriuretic peptide (nesiritide) Jose Iglesias 1,2, Deborah Hom 2, Maria Antoniotti 2, Sammy Ayoub 2 and Jerrold S. Levine 3 1 Nephrology, UMDNJ School of Osteopathic Medicine, Stratford and Robert Wood Johnson School of Medicine, New Brunswick, 2 Departments of Nursing, Pharmacy and Nephrology, Community Medical Center, Toms River, NJ and 3 Nephrology, University of Illinois at Chicago School of Medicine, Chicago, Ill, USA Abstract Background. A recent meta-analysis suggested that the use of nesiritide (NES), a new agent for the treatment of congestive heart failure (CHF), is associated with an increased risk of acute renal failure (ARF). Methods. We examined this issue among 219 consecutive CHF patients, and determined the risk factors for development of ARF [defined as a rise in serum creatinine (SCr) >0.3 mg/dl]. The sole primary outcome was the development of ARF. Results. Seventy one of 219 patients received NES. There was no difference in ARF between patients receiving vs not receiving NES (29 vs 20%, P ¼ 0.17). Evaluation of the entire cohort employing forward stepwise regression analysis revealed the following independent predictors of ARF: admission blood urea nitrogen (BUN) [P ¼ 0.0004, odds ratio (OR) ¼ 1.026], and admission brain natiuretic peptide (P ¼ 0.04, OR ¼ 1.0003). We repeated the same analysis for the subgroups of patients receiving or not receiving NES. For patients not receiving NES (n ¼ 148), ARF developed in 30 (20%), with lower estimated glomerular filtration rate and older age being independent predictors. For patients receiving NES (n ¼ 71), ARF developed in 21 (29%), with hypertension, elevated BUN/SCr ratio, and lack of use of angiotensin inhibitors being independent predictors. Conclusion. Among all patients with CHF, the use of NES was not an independent risk factor for the development of ARF. However, risk factors for developing ARF differed among patients receiving vs not receiving NES. Comparison of these differing factors suggests that administering NES in the setting of diminished renal perfusion and/or altered renal autoregulation may confer an increased risk of ARF. Correspondence and offprint requests to: Jose I. Iglesias, 11 Paulette lane, Howell, NJ 07731, USA. Email: jiglesias@verizon.net Keywords: acute renal failure; brain natriuretic peptide; congestive heart failure; nesiritide Introduction Acute renal failure (ARF) occurs commonly during the course and treatment of acute decompensated congestive heart failure (CHF). Deterioration of renal function in patients with CHF is associated with increased morbidity and mortality, increased rates of rehospitalization and limitations on therapy [1]. The aetiology of ARF in the setting of CHF is complex. Neurohormones like norepinephrine, angiotensin and aldosterone play critical roles in the regulation of both cardiac and renal physiology. In addition, therapeutic agents known to improve clinical outcome and symptoms in CHF, such as angiotensin converting enzyme inhibitors (ACEI), angiotensin-2 receptor antagonists (A2RB), aldosterone inhibitors and loop diuretics, are themselves associated with worsening renal function. Risk factors for the development of ARF with these agents have been previously described in detail [2 4]. Recombinant human B-type brain natriuretic peptide, or brain natriuretic peptide (BNP), is a new agent used in the treatment of CHF [5]. BNP has been given the generic name of neseritide (NES). NES is a potent vasodilator shown to decrease pulmonary artery wedge pressure and relieve pulmonary congestion in patients with CHF [5]. Despite its therapeutic efficacy, NES can adversely affect renal function in susceptible individuals, and cases of ARF have been reported [6]. Two recent meta-analyses of pooled data from five randomized controlled trials have suggested that NES use is associated with an increased risk of both worsening renal function and short-term (<30 day) mortality [6,7]. However, a major limitation of both analyses was a lack of statistical adjustment for ß The Author [2006]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org
Renal failure in CHF with nesiritide 3459 baseline differences between treatment groups other than the assignment of treatment. Indeed, risk factors associated with the development of ARF among CHF patients receiving NES have not been clearly identified. To address the potential role of NES in ARF and to identify potential risk factors for the development of ARF in the setting of NES usage, we took two approaches. First, we examined whether NES is an independent risk factor for ARF among consecutive patients admitted to a community hospital with a diagnosis of acute decompensated CHF (n ¼ 219). To determine which factors associated with the development of ARF among all CHF patients were independent predictors of ARF, we used multivariate analysis. Second, to gain insight as to whether NES might predispose to ARF under specific conditions, we compared the risk factors associated with the development of ARF for patients who received NES (n ¼ 71) vs those who did not (n ¼ 148). To adjust for baseline differences between patients who received NES vs those who did not, we again used multivariate analysis. Methods Patients In order to identify and evaluate the risk factors associated with the development of ARF in acute decompensated CHF patients in general and in such patients receiving NES, we reviewed the clinical course of 219 consecutive patients undergoing therapy for CHF, 71 of whom (32%) received intravenous NES. These patients were admitted between 1 September 2003 and 31 August 2004. NES was administered as an intravenous bolus of 2 mg/kg followed by a continuous infusion at a rate of 0.01 mg/kg/min. Study subjects were identified in one of two ways: (i) Trendstar Õ decision support software and database (McKesson San Francisco, CA, USA) and (ii) Joint Commission of the Accreditation of Health Care Organization (JCAHO)/Center for Medicare Services (CMS), Peer Review Organization of New Jersey (PRONJ), 7th Scope of Work involving CHF [8]. Patients with CHF were identified by the diagnosis-related group (DRG) code 127 and the 9th International Classification of Disease- Clinical Modification (ICD-9-CM) codes (428., 398.91, 402.01, 402.11, 402.91, 404.00, 404.10 13, and 404.90 93) [9]. All DRG and ICD-9-CM codes were entered by professional coders who were approved by the respective agencies. All patients entered into this study had at least one sign or symptom consistent with the diagnosis of CHF, including dyspnoea at rest, dyspnoea with minimal exertion, peripheral oedema, rales or radiological evidence of CHF. Patients under the age of 18 and those with end-stage renal disease on maintenance dialytic therapy were excluded from the analysis. Evaluation included the following admission variables: serum sodium (Na), blood urea nitrogen (BUN), haematocrit (Hct), serum creatinine (SCr), plasma BNP, peak SCr during admission, medications, comorbidities, demographics, documentation of left ventricular ejection fraction <40% and admission physiological variables. Medications, comorbidities and demographic and physiological variables were abstracted from the medical record. ARF was defined as an increase of SCr >0.3 mg/dl occurring any time within 30 days of admission. Our choice of this threshold, while arbitrary, is based upon previous studies showing that an increase in SCr of 0.3 mg/dl is associated with a longer hospital stay and an increased mortality (both in-hospital and long-term) [10]. Moreover, in the recent meta-analysis demonstrating an increased risk of ARF in association with NES, ARF was similarly defined as an increase in SCr >0.5 mg/dl occurring any time within 30 days of NES administration [6]. Glomerular filtration rate (GFR) was estimated from the simplified equation developed by the Modification of Diet in Renal Disease study [11]. Data analysis Primary outcome is the development of ARF occurring within 30 days of hospitalization. The groups we evaluated included the entire cohort (CHF), as well as the sub-cohorts receiving NES (CHF þ NES) and not receiving NES (CHF þ NO NES). Summary statistics were computed for the entire study cohort, as well as for the NES and NO NES sub-cohorts. We performed both univariate and multivariate analyses. Continuous variables were expressed as mean SD, and compared with the Student t-test or the Wilcoxon rank-sum test, as appropriate. Categorical values were compared with Fisher s exact test. Variables which were significant by univariate analysis at P < 0.05 were candidates for multivariate analysis. Logistic regression analysis with forward variable selection was performed to determine variables independently predictive of ARF. In logistic regression, step selections were based on the maximum likelihood ratio. For continuous variables, the odds ratio (OR) represents the relative amount by which the probability of obtaining the outcome variable will increase or decrease if the independent variable is increased by exactly one unit. ORs and their 95% confidence intervals (CIs) were determined by exponentiation of the regression coefficient and its upper and lower 95% CI, respectively. Results Risk factors for ARF in acute decompensated CHF To determine the risk factors that correlate with the development of ARF among patients presenting with acute decompensated CHF, we retrospectively analysed multiple demographic, clinical and laboratory variables, all routinely obtained on admission. We evaluated 219 consecutive patients with acute decompensated CHF, of whom 71 (31%) received NES. ARF developed in 51 of the 219 patients (23%). The use of NES did not correlate with the development of ARF (Table 1), as there was no statistically significant difference in the incidence of ARF between those patients who received NES and those who did not (29 vs 20%, P ¼ 0.17, Fisher s exact test). The duration of treatment with NES did not differ between patients who developed ARF and those who did not (2.9 2.5 days vs 2.7 2.0 days, P ¼ 0.93, Wilcoxon rank-sum test). None of the 51 patients who
3460 J. Iglesias et al. Table 1. Incidence of ARF in CHF patients receiving vs not receiving nesiritide P-value Odds ratio 95% CI All patients (n ¼ 219), ARF defined as SCr > 0.3 mg/dl ARF (n ¼ 51) No ARF (n ¼ 168) NES (n ¼ 71) 21 (29%) 50 (71%) 0.17 1.6 0.86 3.14 NO NES (n ¼ 148) 30 (20%) 118 (80%) Survivors only (n ¼ 177), ARF defined as SCr > 0.3 mg/dl ARF (n ¼ 39) No ARF (n ¼ 138) NES (n ¼ 54) 12 (22%) 42 (78%) 0.96 1.01 0.48 2.2 NO NES (n ¼ 123) 27 (22%) 96 (78%) NES given within 48 h of admission (n ¼ 204), ARF defined as SCr > 0.3 mg/dl ARF (n ¼ 42) No ARF (n ¼ 162) NES (n ¼ 56) 12 (21%) 44 (79%) 0.8 1.07 0.51 2.29 NO NES (n ¼ 148) 30 (20%) 118 (80%) All patients (n ¼ 219), ARF defined as SCr > 0.5 mg/dl ARF (n ¼ 32) No ARF (n ¼ 187) NES (n ¼ 71) 7 (10%) 64 (90%) 0.22 0.54 0.22 2.93 NO NES (n ¼ 148) 25 (17%) 123 (83%) developed ARF required dialysis. ARF significantly increased the length of stay in the hospital. The average length of stay among patients developing ARF was 11.9 9.8 days, as compared with the 7.2 6.8 days for patients without ARF (P ¼ 0.00003, Wilcoxon ranksum test). Limiting the analysis for length of stay to survivors did not affect this result (11.7 10.1 days vs 6.8 5.5 days, P ¼ 0.00014). Since the recent meta-analysis described an increased short-term mortality in association with NES, it is possible that early mortality may have masked the development of ARF, that is, patients may have died before ARF could occur. To address this possibility, we reanalysed our data, including only survivors (>30 days) (Table 1). Even among survivors, NES was not associated with the development of ARF (22 vs 22%, P ¼ 1.0, Fisher s exact test). An additional factor that could have affected our data is the possibility that NES was used as second-line or salvage therapy. However, this does not appear to be the case. Of the 71 patients who received NES, 56 (79%) received NES within 48 h of admission. Moreover, there was no statistical difference in the day of NES administration comparing patients with ARF and those without ARF. When we restricted our analysis to those 56 patients receiving NES within 48 h of admission (Table 1), NES was still not associated with the development of ARF (21 vs 20%, P ¼ 0.8, Fisher s exact test). In the recent meta-analysis that demonstrated an increased risk of ARF in association with NES, ARF was defined as an increase of SCr >0.5 mg/dl. When we reanalysed our data using this stricter definition of ARF, the use of NES still did not correlate with the development of ARF (Table 1). ARF developed in 32 of the 219 patients (15%). There was no statistically significant difference in the incidence of ARF between those patients who received NES and those who did not (10 vs 17%, P ¼ 0.22, Fisher s exact test). Table 2 lists the results of a univariate analysis for all CHF patients whose admission characteristics were associated with ARF. The following demographic, clinical and laboratory variables were significantly associated with the development of ARF: increased age (P ¼ 0.027), increased SCr (P ¼ 0.019), increased serum BNP (P ¼ 0.008), increased BUN (P ¼ 0.00003), increased BUN/SCr ratio (P ¼ 0.0012), decreased estimated GFR (P ¼ 0.003) and a history of hypertension (HTN) (P ¼ 0.037, OR ¼ 2.05, 95% CI ¼ 1.05 4.0). By multivariate regression analysis, the only independent predictors of the development of ARF were (in descending order of coefficient of determination) increased BUN and increased BNP levels (Table 3). Although NES did not achieve statistical significance by univariate analysis, a greater percentage of patients with ARF received NES than did those without ARF (41 vs 30%, P ¼ 0.17, Fisher s exact test). To ensure that NES was not associated with ARF, we included NES in our multivariate model. By multivariate analysis, NES still did not predict the development of ARF (P ¼ 0.42, OR ¼ 0.83, 95% CI ¼ 0.36 1.90). Risk factors for ARF in the presence and absence of NES To determine whether NES usage affected the risk factors predisposing to ARF, we compared the risk factors associated with ARF among those CHF patients who received NES (n ¼ 71) and those who did not receive NES (n ¼ 148). For patients who received NES, by univariate analysis, the following admission characteristics were significantly associated with the development of ARF (Table 4): increased BUN (P ¼ 0.004), increased BUN/SCr ratio (P ¼ 0.013), a history of HTN (P ¼ 0.003, OR ¼ 6.78, 95% CI ¼ 1.76 26.0), lack of use of either an ACEI or A2RB
Renal failure in CHF with nesiritide 3461 Table 2. Univariate analysis of risk factors at admission associated with the development of ARF in all CHF patients Demographic variables No ARF (n ¼ 168) ARF (n ¼ 51) P-value Odds ratio 95% CI Age 79 9.9 82 9.8 0.027 Male sex 85 (51%) 21 (39%) 0.26 0.68 0.36 1.3 Caucasian race 163 (96%) 51 (100%) 0.47 3.46 0.2 16.3 Laboratory variables SCr (mg/dl) 1.43 0.6 1.69 0.7 0.019 Sodium (meq/l) 137 5 137 6 0.55 BNP (pg/ml) 1235 1184 1959 1663 0.008 BUN (mg/dl) 32 23 47 29 0.00003 BUN/SCr 22 10 31 25 0.0012 Haematocrit 35 5.5 35 5.6 0.85 GFR 56 28 43 20 0.003 Haemodynamic and physiological variables MAP (mm/hg) 95 17 94 21 0.22 Ejection fraction 40% 73 (42%) 21 (43%) 0.55 0.79 0.36 1.7 Pulse pressure 64.9 24 64.7 23 0.96 Comorbidities Hypertension 90 (54%) 36 (70%) 0.037 2.05 1.05 4.0 Coronary artery disease 121 (72%) 32 (63%) 0.22 0.64 0.33 1.24 COPD 73 (43%) 22 (43%) 1 0.97 0.52 1.8 Diabetes 63 (37%) 18 (35%) 0.87 0.9 0.47 1.73 Atrial fibrillation 72 (43%) 18 (35%) 0.33 0.72 0.38 1.38 Medications Inotropes 20 (12%) 12 (24%) 0.07 2.26 1.02 5.0 Diuretics 159 (95%) 50 (98%) 0.69 2.52 0.31 20.6 ACEI/A2RB 93 (55%) 27 (53%) 0.75 0.9 0.48 1.68 Nitrates 104 (62%) 34 (67%) 0.62 1.21 0.62 2.35 Beta blocker 108 (64%) 35 (69%) 0.74 1.19 0.61 2.34 Digoxin 65 (38%) 18 (35%) 0.74 0.86 0.45 1.64 Nesiritide 50 (30%) 21 (41%) 0.17 1.65 0.86 3.16 ICU admission 33 (20%) 15 (29%) 0.17 1.70 0.84 3.48 A2RB, angiotensin-2 receptor blocker; ACEI, angiotensin converting enzyme inhibitor; BNP, brain natiuretic peptide; BUN/SCr, blood urea nitrogen to serum creatinine ratio; COPD, chronic obstructive pulmonary disease; GFR, glomerular filtration rate; ICU, intensive care unit; MAP, mean arterial pressure; SCr, serum creatinine. Values are expressed as mean SD or number (percent). Odds ratios and 95% CI are presented only for discrete binary variables. Conversion factors for SI units: SCr: mg/dl 88.4 ¼ mmol/l; BUN: mg/dl 0.375 ¼ mmol/l. Table 3. Multivariate analysis of risk factors at admission associated with the development of ARF Regression coefficient SE Chi-square Coefficient of determination P-value Odds ratio 95% CI Entire cohort BUN 0.018 0.0063 8.53 0.034 0.003 1.018 1.006 1.030 BNP 0.0003 0.00012 5.51 0.024 0.018 1.0003 1.00007 1.00053 NES Hypertension 1.87 0.74 6.4 0.099 0.011 6.47 1.52 27 ACEI/A2RB 1.38 0.63 5.4 0.067 0.028 0.25 0.072 0.86 BUN/SCr 0.057 0.026 4.7 0.062 0.035 1.057 1.03 1.10 NO NES GFR 0.02 0.01 4.22 0.034 0.024 0.98 0.96 0.99 Age 0.058 0.029 5.07 0.028 0.028 1.058 1.008 1.10 To confirm the results of univariate analysis showing that NES was not associated with ARF, we included NES in our multivariate analysis. On multivariate analysis, NES was still not a predictor of ARF (coefficient of determination, 0.0009; P ¼ 0.42; OR ¼ 0.83, 95% CI, 0.36 1.90). (P ¼ 0.01, OR ¼ 2.75, 95% CI ¼ 1.34 5.51). Of these, by multivariate analysis, the independent predictors of ARF were determined (in descending order of coefficient of determination): history of HTN, lack of ACEI or A2RB usage and increased BUN/SCr ratio (Table 3). For CHF patients who did not receive NES, by univariate analysis, the following characteristics were significantly associated with the development of ARF (Table 5): older age (P ¼ 0.010), increased BUN (P ¼ 0.004), decreased estimated GFR (P ¼ 0.02) and the absence of a history of coronary artery disease (P ¼ 0.045). Of these, decreased estimated GFR and increased age were independent predictors of ARF (Table 3).
3462 J. Iglesias et al. Table 4. Univariate analysis of risk factors at admission associated with the development of ARF in CHF patients receiving nesiritide Demographic variables No ARF (n ¼ 50) ARF (n ¼ 21) P-value Odds ratio 95% CI Age 77 11.8 79 11.6 0.50 Male sex 24 (48%) 10 (48%) 0.97 0.98 0.35 2.73 Caucasian race 49 (96%) 21 (100%) 0.88 0.45 0.007 25 Laboratory variables SCr (mg/dl) 1.7 0.73 1.9 0.74 0.17 Sodium (meq/l) 136 4 134 3.8 0.20 BNP (pg/ml) 2063 1407 2548 1666 0.28 Blood urea nitrogen (mg/dl) 43 31 63 32 0.004 BUN/SCr 23 10.2 34.9 20 0.013 Haematocrit 35 5.4 33 4.8 0.10 GFR 43 20 37 19 0.14 Haemodynamic and physiological variables MAP (mm/hg) 93 18.7 94 23 0.90 Ejection fraction 40% 20 (40%) 8 (38%) 0.50 0.60 0.16 2.24 Pulse pressure 62 30 69 24 0.40 Comorbidities Hypertension 23 (46%) 18 (86%) 0.003 6.78 1.76 26 Coronary artery disease 39 (78%) 17 (81%) 1 1.09 0.3 3.96 COPD 15 (30%) 8 (38%) 0.59 1.39 0.48 4.1 Diabetes 22 (44%) 10 (48%) 1 1.11 0.40 3.1 Atrial fibrillation 18 (36%) 5 (24%) 0.41 0.54 0.17 1.7 Medications Inotropes 17 (34%) 9 (43%) 0.59 1.412 0.49 4.02 Diuretics 46 (92%) 20 (95%) 1 1.30 0.13 13 ACEI/A2RB 38 (76%) 9 (43%) 001 2.75 1.34 5.51 Nitrates 35 (70%) 14 (67%) 0.77 0.80 0.27 2.40 Beta blocker 31 (62%) 15 (71%) 0.59 1.45 0.48 4.41 Digoxin 23 (46%) 8 (38%) 0.60 0.70 0.25 1.98 ICU admission 17 (34%) 12 (57%) 0.11 2.59 0.91 7.35 Table 5. Univariate analysis of risk factors at admission associated with the development of ARF in CHF patients not receiving nesiritide Demographic variables o ARF (n ¼ 118) ARF (n ¼ 30) P-value Odds ratio 95% CI Age 80 8.9 84 7.8 0.010 Male sex 61 (52%) 11 (37%) 0.16 0.54 0.2 1.23 Caucasian race 114 (97%) 30 (100%) 0.70 0.40 0.008 0.20 Laboratory variables SCr (mg/dl) 1.31 0.63 1.51 0.7 0.1 Sodium (meq/l) 136 4.7 138 6.8 0.53 BNP (pg/ml) 918 893 1518 1545 0.088 Blood urea nitrogen (mg/dl) 27.3 16 36 20 0.004 BUN/SCr 21 10 28 28 0.054 Haematocrit 35 5.7 36 5.8 0.30 GFR 60 29 46 19 0.02 Haemodynamic and physiological variables MAP (mm/hg) 96 16 93 19 0.35 Ejection fraction 40% 53 (45%) 13 (43%) 0.80 0.87 0.33 2.27 Pulse pressure 66 20 61 0.13 0.11 Comorbidities Hypertension 67 (57%) 18 (60%) 0.83 1.14 0.50 2.58 Coronary artery disease 82 (69%) 15 (50%) 0.045 0.43 0.19 0.99 COPD 58 (49%) 14 (47%) 0.84 0.905 0.41 2.02 Diabetes 41 (35%) 8 (27%) 0.51 0.638 0.27 1.66 Atrial fibrillation 54 (46%) 13 (43%) 0.84 0.91 0.40 2.00 Medications Inotropes 3 (2.5%) 3 (10%) 0.098 4.25 0.81 22.3 Diuretics 113 (96%) 30 (100%) 0.58 1.26 1.16 1.38 ACEI/A2RB 55 (47%) 18 (60%) 0.22 1.71 0.76 3.88 Nitrates 69 (58%) 20 (67%) 0.53 1.42 0.61 3.3 Beta blocker 77 (65%) 20 (67%) 1 1.06 0.46 2.49 Digoxin 42 (35%) 10 (33%) 1 0.9 0.39 2.11 ICU admission 16 (13%) 3 (10%) 0.76 0.71 0.19 2.6
Renal failure in CHF with nesiritide 3463 Discussion Among patients hospitalized for CHF, a decline in renal function is associated with a poor prognosis [1,10,12,13]. In addition to portending a poor prognosis, renal dysfunction can interfere with and limit the effective treatment of CHF [2,10,13,14]. Although the aetiology of renal dysfunction in patients with CHF is complex, an imbalance between renal vasodilator and vasoconstrictive influences plays a key role. Of the neurohumoral vasodilators, one of the more prominent is the BNP [5,15]. NES, the generic name given to human recombinant BNP, has several pharmacological effects that are desirable in the setting of CHF. Despite the potential benefits of NES in patients with CHF, there is reason for caution. In some reports, NES has been associated with a decline in renal function, especially when administered in doses >0.015 mg/kg/ min [6,16,17]. Moreover, a recent meta-analysis of five randomized, double-blind parallel group trials of NES in CHF patients (n ¼ 1269) reported an increased frequency of renal impairment among patients receiving NES (11 vs 4%; relative risk 1.54; 95% CI ¼ 1.19 1.98; P ¼ 0.03)[6]. However, a major limitation of this meta-analysis is that there were inadequate data to control for any confounding differences beyond treatment group assignments. To explore the potential role of NES in ARF, we used two complementary risk-adjusted approaches. First, we determined whether NES usage was a risk factor for the development of ARF, in 219 consecutive patients admitted with the diagnosis of acute decompensated CHF. Second, since NES may alter the underlying renal physiology, we compared the risk factors for ARF in CHF patients who received NES and those who did not. In contrast to a recently published meta-analysis, NES did not emerge as a risk factor for ARF, even upon univariate analysis. Among over 25 demographic, clinical and laboratory variables, only elevated BUN and elevated BNP at admission independently predicted ARF. Our results are in accord with several studies suggesting that among CHF patients BUN may reflect the underlying renal function better than SCr [18,19]. This is consistent with the notion that for patients with CHF the principal determinant of renal function is the renal blood flow, and declines in renal blood flow typically affect the clearance of BUN disproportionately more than they do the clearance of creatinine [19]. It is likely that elevated BNP levels were a marker of worsening CHF and a consequent decrease in renal plasma flow. Overall, our data are in agreement with earlier studies of risk factors for ARF in CHF patients, in which the markers of pre-existing renal impairment and/or severity of CHF were strong predictors of ARF [12,14,20]. When we analysed the risk factors for ARF in the presence vs absence of NES, we again found that preexisting renal insufficiency was an independent risk factor, irrespective of NES usage. For patients receiving NES, an increased BUN/SCr ratio at admission, consistent with compromised renal perfusion, predicted ARF. For these patients, a history of HTN and a lack of ACEI or A2RB usage were also independent predictors of ARF. We speculate that patients who received NES and developed ARF may have previously demonstrated worsened renal function during inhibition of the renin angiotensin aldosterone axis. These patients may be critically dependent on angiotensin for the maintenance of GFR and therefore predisposed to ARF with vasodilators like NES, which inhibit the release of renin. Patients with a history of hypertension may also be predisposed to ARF during NES therapy because of altered renal autoregulation as a result of underlying hypertensive arteriolosclerosis. Thus, NES may predispose to the development of ARF in the setting of critically diminished renal perfusion and/or altered renal autoregulation. For patients not receiving NES, in accord with other studies, decreased estimated GFR and older age emerged as independent predictors of ARF [4,14,20]. It is noteworthy that the predictors of ARF differed in the presence vs absence of NES, suggesting that the underlying renal haemodynamics may differ in patients receiving NES, either because of the selection or perhaps because of the effects of NES itself on renal physiology. While the risk factors identified for patients receiving NES correlated with diminished renal blood flow and altered autoregulation, those for patients not receiving NES were independent of haemodynamic factors. Thus, decreased estimated GFR emerged as the marker of pre-existing renal insufficiency for patients not receiving NES, as opposed to an elevated BUN/SCr ratio for patients receiving NES. Two known risk factors for ARF, especially in the setting of CHF, non-steroidal anti-inflammatory drugs (NSAIDs) and radiographic contrast dye, do not seem to have played a role in our study. No patient in our study was receiving NSAID at the time of admission, nor was any patient administered NSAID while in the hospital. We cannot exclude the possibility that patients took NSAIDs in the period preceding hospitalization. Also, since the hospital was not approved for acute coronary intervention at the time of our study, no patient received an acute angiographic procedure. There are several limitations to our present study. First, because our study is retrospective, we were limited in our ability to adjust for differences in baseline characteristics between patients receiving or not receiving NES. Indeed, patients receiving NES differed significantly in all of the following baseline characteristics, indicative of more severe CHF and illness: increased SCr, BUN and BUN/SCr ratio; decreased estimated GFR; lower serum Na; increased prevalence of coronary artery disease and chronic obstructive pulmonary disease and increased rate of admission to the ICU (Table 6). To adjust for these baseline differences, we used multivariate analysis.
3464 J. Iglesias et al. Table 6. Baseline characteristics at admission of CHF patients receiving versus not receiving nesiritide Demographic variables NO NES (n ¼ 148) NES (n ¼ 71) P-value Odds ratio 95% CI Age 81 8.9 78 11.7 0.055 Male sex 72 (49%) 34 (48%) 1 0.97 0.55 1.70 Caucasian race 144 (97%) 70 (97%) 0.69 1.04 0.29 5.56 Laboratory variables SCr (mg/dl) 1.35 0.6 1.79 0.7 <0.00001 Sodium (meq/l) 137 5.2 135 3.9 0.005 BNP (pg/ml) 1042 1082 2222 1501 <0.00001 BUN (mg/dl) 29 17 49 32 <0.00001 BUN/SCr 22 15 27 14 0.014 Haematocrit 35 7.5 34.8 5.3 0.59 GFR 58 28 42 20 0.00001 Haemodynamic and physiological variables MAP (mm/hg) 95 15 94 19 0.5 Ejection fraction 40% 66 (45%) 28 (39%) 0.71 1.16 0.55 2.43 Pulse pressure 65 21 64 28 0.62 Comorbidities Hypertension 85 (57%) 41 (58%) 0.88 1.05 0.59 1.86 Coronary artery disease 97 (66%) 56 (79%) 0.039 2.10 1.07 4.13 COPD 72 (49%) 23 (32%) 0.029 0.52 0.28 0.93 Diabetes 49 (33%) 32 (45%) 0.098 1.70 0.95 3.00 Atrial fibrillation 67 (45%) 23 (32%) 0.10 0.59 0.33 1.07 Medications Inotropes 6 (4%) 26 (36%) 0.00001 13.0 5.4 36 Diuretics 143 (97%) 66 (93%) 0.47 0.58 0.15 2.22 ACEI/A2RB 73 (49%) 47 (66%) 0.020 2.00 1.15 3.80 Nitrates 89 (60%) 49 (69%) 0.177 1.55 0.84 2.84 Beta blocker 97 (65%) 46 (65%) 1 1.008 0.55 1.83 Digoxin 52 (35%) 31 (44%) 0.23 1.47 0.82 2.62 ICU admission 19 (13%) 29 (41%) 0.00001 4.68 2.39 9.21 A2RB, angiotensin-2 receptor blocker; ACEI, angiotensin converting enzyme inhibitor; BNP, brain natiuretic peptide; BUN/SCr, blood urea nitrogen to serum creatinine ratio; COPD, chronic obstructive pulmonary disease; GFR, glomerular filtration rate; ICU, intensive care unit; MAP, mean arterial pressure; SCr, serum creatinine. Values are expressed as mean SD or number (percent). Odds ratios and 95% CI are presented only for discrete binary variables. Conversion factors for SI units: SCr: mg/dl 88.4 ¼ mmol/l; BUN: mg/dl 0.375 ¼ mmol/l. Importantly, despite the fact that patients receiving NES tended to be sicker and have more severe CHF, the use of NES was still not associated with an increased risk of ARF within the entire cohort of CHF patients. Thus, even in the face of overall poorer health and more severe CHF, NES usage still did not predict the development of ARF. These data suggest that the use of NES per se may not predispose to the development of ARF. Second, we lacked sufficient data to adjust for unmeasured potential baseline confounders, such as pre-existing structural renal abnormalities, degree of proteinuria, levels of C-reactive protein, volume status, cardiac output, specific measures of organ perfusion and other factors that may be important to the risk for developing ARF. We also lacked sufficient data to adjust for clinical variables indicative of therapeutic response, such as diuretic dose or urine output, which likely affected the clinical decision whether or not to use NES. In this regard, it is important to note that our data do not support the idea that NES was used as a salvage therapy in CHF patients who failed to respond to other therapies or whose clinical status was deteriorating (Table 1). Finally, our definition of ARF as an increase in SCr of 0.3 mg/dl or greater, while in accord with that used by multiple other studies, does not permit us to differentiate renal from pre-renal causes of ARF [10]. It should be noted that the definition we used is more sensitive than that used in the recent meta-analysis of NES and ARF (SCr 0.5 mg/dl). We cannot exclude the possibility that the lack of an association between NES and ARF is the result of a beta-type error, and that a larger study population would have revealed a statistically significant association. In summary, we have examined the risk factors for ARF among CHF patients, both in the presence and in the absence of NES. A major strength of our study is that our cohort of patients is derived from a large community hospital and is therefore clinically and demographically similar to CHF patients within the general population. An important finding of our study is that within the entire cohort of CHF patients, the use of NES did not emerge as an independent risk factor for the development of ARF. However, when we compared the risk factors associated with the development of ARF for patients who received vs not received NES,
Renal failure in CHF with nesiritide 3465 we found different independent risk factors for the two groups (Table 3). This suggests that NES may increase the risk of ARF in selected subsets of patients with CHF. Based upon a comparison of the risk factors for these two groups, we hypothesize that the risk for ARF in the setting of NES may be heightened in states of diminished renal perfusion and/or impaired renal autoregulation. This possibility should be addressed in future prospective studies. Acknowledgements:. This work was supported by NIH grants DK59793 and HL69722, and by a Young Investigator Award from the National Kidney Foundation of Illinois. Conflict of interest statement. Dr Jose Iglesias is on the bureau of SCIOS speakers. References 1. Gottlieb SS, Abraham W, Butler J et al. The prognostic importance of different definitions of worsening renal function in congestive heart failure. J Card Fail 2002; 8: 136 141 2. Mandal AK, Markert RJ, Saklayen MG, Mankus RA, Yokokawa K. Diuretics potentiate angiotensin converting enzyme inhibitor-induced acute renal failure. Clin Nephrol 1994; 42: 170 174 3. Knight EL, Glynn RJ, McIntyre KM, Mogun H, Avorn J. Predictors of decreased renal function in patients with heart failure during angiotensin-converting enzyme inhibitor therapy: results from the studies of left ventricular dysfunction (SOLVD). Am Heart J 1999; 138: 849 855 4. Svensson M, Gustafsson F, Galatius S, Hildebrandt PR, Atar D. How prevalent is hyperkalemia and renal dysfunction during treatment with spironolactone in patients with congestive heart failure? J Card Fail 2004; 10: 297 303 5. Colucci WS, Elkayam U, Horton DP et al. Intravenous nesiritide, a natriuretic peptide, in the treatment of decompensated congestive heart failure. Nesiritide study group. N Engl J of Med 2000; 343: 246 253 6. Sackner-Bernstein JD, Skopicki HA, Aaronson KD. Risk of worsening renal function with nesiritide in patients with acutely decompensated heart failure. Circulation 2005; 11: 1487 1491 7. Sackner-Bernstein JD, Kowalski M, Fox M, Aaronson K. Shortterm risk of death after treatment with nesiritide for decompensated heart failure: a pooled analysis of randomized controlled trials. JAMA 2005; 293: 1900 1905 8. Centers for medicare and medicaid services national heart failure quality improvement project. United States Department Health and Human Services, Baltimore, MD: 2001; 1 13 9. US Health Care Financing Administration. Diagnosis-related groups definitions manual. Wallingford CT, 3M Health Information Services: 1993 10. Butler J, Forman DE, Abraham WT et al. Relationship between heart failure treatment and development of worsening renal function among hospitalized patients. Am Heart J 2004; 147: 331 338 11. Rule AD, Larson TS, Bergstralh EJ, Slezak JM, Jacobsen SJ, Cosio FG. Using serum creatinine to estimate glomerular filtration rate: accuracy in good health and in chronic kidney disease. Ann Intern Med 2004; 141: 929 937 12. Forman DE, Butler J, Wang Y et al. Incidence, predictors at admission, and impact of worsening renal function among patients hospitalized with heart failure. J Am Coll Cardiol 2004; 43: 61 67 13. Heywood JT. The cardiorenal syndrome: lessons from the ADHERE database and treatment options. Heart Fail Rev 2004; 9: 195 201 14. Kittleson M, Hurwitz S, Shah MR et al. Development of circulatory-renal limitations to angiotensin-converting enzyme inhibitors identifies patients with severe heart failure and early mortality. J Am Coll Cardiol 2003; 41: 2029 2035 15. Levin ER, Gardner DG, Samson WK. Natriuretic peptides. N Engl J Med 1998; 339: 321 328 16. Burger AJ, Horton DP, LeJemtel T et al. Effect of nesiritide (B-type natriuretic peptide) and dobutamine on ventricular arrhythmias in the treatment of patients with acutely decompensated congestive heart failure: the PRECEDENT study. Am Heart J 2002; 144: 1102 1108 17. Abraham W. Serum creatinine elevations in patients receiving nesiritide are related to starting dose. Circulation 2005; 112 [Suppl 17]: S-589 18. Ljungman S, Laragh JH, Cody RJ. Role of the kidney in congestive heart failure. Relationship of cardiac index to kidney function. Drugs 1990; 39 [Suppl 4]: 10 21 19. Cody RJ, Ljungman S, Covit AB et al. Regulation of glomerular filtration rate in chronic congestive heart failure patients. Kidney Int 1988; 34: 361 367 20. Weinfeld MS, Chertow GM, Stevenson LW. Aggravated renal dysfunction during intensive therapy for advanced chronic heart failure. Am Heart J 1999; 138: 285 290 Received for publication: 11.4.06 Accepted in revised form: 22.6.06