Pediatric Nephrology

, Volume 26, Issue 10, pp 1909–1912

Intensive hemodialysis for cardiomyopathy associated with end-stage renal disease


    • Division of Nephrology and HypertensionCincinnati Children’s Hospital Medical Center
  • Jens Goebel
    • Division of Nephrology and HypertensionCincinnati Children’s Hospital Medical Center
  • Mark Mitsnefes
    • Division of Nephrology and HypertensionCincinnati Children’s Hospital Medical Center
  • Angela Lorts
    • Division of CardiologyCincinnati Children’s Hospital Medical Center
  • Benjamin Laskin
    • Division of Nephrology and HypertensionCincinnati Children’s Hospital Medical Center
Brief Report

DOI: 10.1007/s00467-011-1921-y

Cite this article as:
Nehus, E., Goebel, J., Mitsnefes, M. et al. Pediatr Nephrol (2011) 26: 1909. doi:10.1007/s00467-011-1921-y


Heart and kidney dysfunction often coexist, and increasing evidence supports the interaction of these two organs, as demonstrated by the clinical condition known as cardiorenal syndrome (CRS). We report a pediatric patient with end-stage renal disease (ESRD) who developed a dilated cardiomyopathy and decompensated heart failure after undergoing unilateral nephrectomy and while on maintenance peritoneal dialysis. He showed marked improvement in his cardiac function with the addition of intensive hemodialysis. We discuss the pathophysiology of cardiorenal syndrome in patients with ESRD and suggest that intensive dialysis may be an effective therapy for this condition.


Cardiorenal syndromeUremiaChronic kidney diseaseCardiovascular diseaseEchocardiogram


Cardiorenal syndrome (CRS) is a pathophysiologic state in which the dysfunction of one organ—either the heart or the kidney— causes dysfunction of the other. While CRS initially referred to kidney disease secondary to cardiac dysfunction, it has been redefined to include chronic kidney disease (CKD) leading to cardiac dysfunction. This broadened definition underscores the complex interaction between these two organs, both of which play complementary roles in blood pressure (BP) regulation, diuresis, and natriuresis [1].

CKD is a known risk factor for cardiovascular disease. In addition to hypertension, dyslipidemia, and diabetes, other conditions present in CKD associated with cardiovascular disease include anemia, inflammation, increased calcium/phosphorus product, and elevated levels of homocysteine and asymmetric dimethyl arginine. Furthermore, other uremic toxins may contribute to cardiovascular morbidity in patients with end-stage renal disease (ESRD) [2]. The ability of dialysis to decrease levels of potential mediators of CRS is ill-defined [3]. We report here a pediatric patient with ESRD who developed CRS that improved with intensive hemodialysis (HD).

Case report

An 11-year-old boy on maintenance peritoneal dialysis (PD) for ESRD secondary to focal segmental glomerulosclerosis presented with orthopnea and dyspnea on exertion. Nephrotic syndrome, diagnosed at age 3 years, had become refractory to corticosteroids, mycophenolate mofetil, and cyclosporine. Five months prior to presentation, frequent infections, malnutrition, and progression of CKD prompted unilateral nephrectomy and initiation of PD. His estimated cystatin C-based glomerular filtration rate (GFR) was 11 mL/min/1.73 m2, and his echocardiogram showed normal cardiac dimensions and systolic function at the time of PD initiation. His daily PD prescription was six cycles over 10 h with a mixture of 1.5 and 2.5% dextrose, a dwell volume of 1200 ml (37 mL/kg or 1040 mL/m2), and a last fill of 400 mL.

Physical examination revealed bibasilar rales, an S3 gallop, and pulsus alternans. His heart rate was 112 beats/min; BP, 133/90 mmHg; respiratory rate, 32 breaths/min; room air oxygen saturation, 97%. He had no peripheral edema. His weight and height were 34 kg and 1.43 m, respectively, with a corresponding body mass index (BMI) of 16.7 kg/m2 (32nd percentile). An echocardiogram showed diffuse chamber enlargement, severely depressed biventricular function, and a shortening fraction (SF) of 10% (normal >30%). Laboratory results included a blood urea nitrogen level of 66 mg/dL; serum creatinine, 12.3 mg/dL; calcium, 7.6 mg/dL; phosphorus, 7.7 mg/dL; albumin, 2.2 g/dL; hemoglobin, 7.4 g/dL. Eleven days prior to presentation, his intact parathyroid hormone level was 388 pg/mL (normal 17–73 pg/mL), and iron studies showed a transferrin saturation of 10% and a serum ferritin of 122 ng/mL. At that time, his hemoglobin was 10.4 g/dL and his normalized protein catabolic rate was 1.62 g/kg/day. His estimated GFR based on 24-h urinary creatinine excretion was only 0.5 mL/min/1.73 m2, and his urine output had decreased to about 50 mL per day. On admission, further blood tests showed a brain natriuretic peptide level of 15 166 pg/mL (normal <100 pg/mL); creatinine phosphokinase, 106 U/L (normal 60–370 U/L); troponin I, 0.24 ng/mL (normal <0.04 ng/mL). A nasal swab PCR was positive for human metapneumovirus. Blood PCR studies for adenovirus, enterovirus, and parvovirus, and a nasal swab PCR for influenza and respiratory syncytial virus were negative.

His hospital course is summarized in Fig. 1. He was transfused with packed red blood cells, and inotropic support was initiated with a milrinone infusion. Due to marginal cardiac output, a continuous calcium infusion was added to the regimen. Cardiac magnetic resonance imaging (MRI) performed on hospital day 7 revealed severe left ventricular (LV) dilatation, global myocardial dysfunction, and no T2 evidence of myocarditis. Inotropic support was weaned off by hospital day 8, and the patient’s heart failure was treated with fluid management, angiotensin converting enzyme (ACE) inhibition, and beta blockade. Worsening azotemia, despite an increase in the number of PD cycles, necessitated two HD treatments on hospital days 17 and 18. A midday exchange was added to his PD prescription, and the fill volume was increased to 1400 mL. Adequate PD clearance was subsequently ascertained, with a measured Kt/V of 3.4. His PD prescription at this time was nine cycles over 12 h, a fill volume of 1400 mL, a last fill of 1000 mL, a midday exchange of 1400 mL, and a mixture of 1.5 and 2.5% dextrose. A peritoneal equilibration test showed low average solute transport characteristics. He was discharged on hospital day 24, only to be re-admitted 1 day later after a syncopal episode. His BP at admittance was 61/31 mmHg, pulsus alternans was again present, and an echocardiogram showed severe biventricular dysfunction (SF 12%). Beta blocker therapy was discontinued, and inotropic support was resumed, but could not be weaned due to low cardiac output and hypotension.
Fig. 1

Clinical summary of blood pressure, weight, echocardiogram shortening fraction, and days on inotropic support These graphs represent the clinical course of our patient from presentation until discharge. The lower graph displays his daily weight, only on a larger scale than is used in the upper graph. The average daily blood pressures and cardiac function showed sustained improvement after initiation of hemodialysis (arrows) despite eventual discontinuation of inotropic support. These improvements were noted despite no significant decline in the patient’s weight, excluding improvements in fluid overload as an explanation for these findings

At this time, an LV assist device was considered as a bridge to combined cardiac/renal transplantation. As a last effort, intensive daily HD was initiated to clear any “uremic toxins” possibly contributing to his cardiac dysfunction. Initially, HD (3 h per treatment with a blood flow of 200 mL/min) was performed for clearance only, and ultrafiltration was exclusively achieved with nightly PD. After the patient had demonstrated hemodynamic tolerance of HD, modest ultrafiltration (500–1000 ml per treatment) would occasionally be added to his HD prescription to maintain his dry weight; the PD prescription was left unchanged. Five days later, a repeat echocardiogram showed dramatic improvement of systolic function and a SF of 32%. Discontinuation of PD and inotropic support occurred 10 and 15 days after HD initiation, respectively. The patient remained on HD six times weekly for 4 weeks and was then transitioned to HD treatments three times weekly. His echocardiograms continued to improve, although mild systolic dysfunction remained (Fig. 1).

Aside from occasional hemodynamic intolerance of ultrafiltration, the patient remained asymptomatic. To facilitate gradual fluid removal, daily PD was resumed along with three times weekly HD. At discharge, an echocardiogram was reassuring with a SF of 30%. The patient continued outpatient PD and HD for the next several months and remained asymptomatic. Eight months after discharge, he received a kidney transplant from a living-related donor. His most recent echocardiogram showed normal systolic function and chamber dimensions.


Cardiovascular disease is the leading cause of morbidity and mortality in patients with ESRD [4]. LV hypertrophy is the most common cardiac abnormality in children with ESRD [5], and a significant subset of adult ESRD patients have systolic dysfunction [6]. While reduced systolic function may represent a maladaptive response to hypertension and volume overload, evidence suggests that the uremic milieu itself also has a cardiodepressant effect. Histological studies comparing controls not requiring dialysis and dialyzed patients with dilated cardiomyopathy demonstrate myocyte hypertrophy and increased interstitial fibrosis in the latter [7]. Furthermore, uremic serum can reduce myocyte contractility in vitro [8]. Recent recognition of “organ crosstalk” has focused on alternative mediators, or uremic toxins, that may be involved in the association of cardiac dysfunction with chronic renal insufficiency [2].

In our patient, classical risk factors for myocardial dysfunction—hypertension and volume overload—were not felt to significantly contribute to our patient’s cardiac disease. Although he had a history of hypertension before developing cardiac symptoms, he had a normal echocardiogram just months prior to presentation. Furthermore, his cardiac dimensions at the time of admission did not show increased wall thickness consistent with hypertensive cardiomyopathy. Increased extracellular volume in association with LV hypertrophy has been demonstrated in patients on PD [9]. However, we do not feel that this was responsible for our patient’s cardiac dysfunction because: (1) his weight was stable over his initial 5-month course of PD, (2) attempts of additional volume removal did not improve his cardiac function and only led to further hemodynamic instability, and (3) initiation of HD caused significant improvement of his cardiac function without any reduction in his weight. In fact, his dry weight slightly increased after HD initiation, likely representing an increased tolerance of higher preload due to improved myocardial contractility (Fig. 1).

Our patient demonstrated low hemoglobin and serum albumin levels which may have contributed to his cardiac dysfunction. Metteucci et al. [10] reviewed LV geometry in children with CKD and demonstrated that an increased LV mass index was associated with low hemoglobin, young age, high BMI, and low GFR. They also demonstrated a positive association between eccentric hypertrophy and low serum albumin. However, our patient did not have long-standing anemia (hemoglobin was 10.4 g/dL 11 days prior to presentation), and correction of his anemia at presentation did not result in any immediate improvement of his cardiac function. Furthermore, his serum albumin levels increased after unilateral nephrectomy (at which time he had a normal echocardiogram), but still remained low (2.0–2.5 mg/dL) due to ongoing urinary protein losses and despite high-dose ACE inhibition.

In the absence of an endomyocardial biopsy, we cannot exclude the possibility that our patient spontaneously recovered from a viral myocarditis. However, we are unaware of any reports of human metapneumovirus causing cardiac dysfunction. Furthermore, cardiac magnetic resonance, an accurate diagnostic test in the evaluation of myocarditis [11], did not support this etiology.

Therefore, we hypothesize that uremia per se was largely responsible for our patient’s cardiac dysfunction. To what extent HD may mitigate the accumulation of uremic toxins, in particular those larger than 1 kDa that are more effectively cleared by modern high-flux dialyzers, is still under investigation [3]. Nixon et al. [12] demonstrated an acute increase in contractility following HD evidenced by an upward shift in the Frank–Starling curve, and long-term improvement in ejection fraction has been observed in ESRD patients after the initiation of HD [13]. More recently, Chan et al. [14] demonstrated improved systolic function after conversion from conventional HD (3 days/week, 4 h/session) to nocturnal HD (6 nights/week, 8–10 h/session). Furthermore, color tissue velocity imaging, which permits a more objective and sensitive measurement of cardiac function, has demonstrated impaired LV contractility in patients with ESRD that improved following HD [15].

In our patient, echocardiographic improvement of systolic function was apparent just 5 days after he started HD. Furthermore, indices of cardiac output, including his BP, were noted to increase the day HD was initiated and remained stable while inotropic support was weaned (Fig. 1). Therefore, we feel that HD acutely improved cardiac output, which was sustained with maintenance HD.

In summary, we present a patient with ESRD who developed CRS and severe biventricular systolic dysfunction which improved with frequent HD. We postulate that HD, through a more effective clearance of cardiotoxic uremic solutes, improved cardiac contractility and systolic function. Our patient continued to show improvement in cardiac function after transplantation, which has been reported in other patients [16]. Although kidney transplantation remains the definitive treatment, intensive HD should be considered as a therapeutic option to improve cardiac function in ESRD patients with CRS.

Financial support

No financial support was received.

Financial disclosure

There are no conflicts of interest to disclose.

Copyright information

© IPNA 2011