Managing newborns with kidney failure is a complex undertaking; even under ideal circumstances, dialysis is technically challenging and available therapies are designed for adults. These issues are exacerbated in smaller newborns, and intervention has traditionally not been offered in those below a certain weight threshold. Ethical concerns abound and patients deemed too small for dialysis are typically transitioned to comfort or palliative care. However, many of these neonates are otherwise healthy and would be considered survivable if kidney replacement therapy were available. To challenge the existing paradigm, we present 7 preterm, low birth weight neonates with end-stage kidney disease who were successfully managed using an innovative approach to kidney replacement therapy. These newborns had a median gestational age of 32 weeks (interquartile range [IQR], 32-35) and a median birth weight of 1.58 kg (IQR, 1.41-2.01). Kidney replacement therapy was initiated at a median age of 16 days (IQR, 1.5-40) and a weight of 1.85 kg (IQR, 1.57-2.1). Five of the 7 newborns (71%) survived to hospital discharge. Kidney replacement therapy was provided using 3F and 4F single lumen catheters and a modified ultrafiltration device. Patients experienced excellent metabolic control, and fluid homeostasis was achieved in the first week of life. Furthermore, survivors experienced physiologic weight gain and linear growth throughout their hospitalization. These findings, although preliminary, are encouraging for our smallest patients with kidney failure and suggest that survivability thresholds should be reexamined. At a minimum, neonatologists should be aware that novel approaches exist and may be considered for these challenging patients.
Neonatal care has evolved considerably over the past 30 years, and we are now seeing younger and smaller newborns survive to hospital discharge.1 These patients are tenuous, often experience dysfunction of multiple organ systems, and require complex care regimens. With increasingly preterm gestational age, our ability to provide support is limited by existing techniques and technologies.
Kidney disease is a common complication in preterm neonates. Nephrogenesis is incomplete until the third trimester and reduced nephron mass predisposes to kidney dysfunction.2 Furthermore, systemic disease and the treatments required can cause acute kidney injury (AKI)3 ; indeed, AKI occurs in nearly 50% of neonates born between 22 and 29 weeks’ gestation.4 In severe cases of AKI and, more commonly, in patients with congenitally acquired end-stage kidney disease (ESKD), kidney replacement therapy (KRT) is commonly required.
KRT has traditionally presented technical challenges in preterm neonates. The cornerstone of KRT in neonates, peritoneal dialysis, can be performed in neonates weighing <2 kg; however, peritoneal dialysis catheter insertion is challenging, particularly if the patient is edematous or malnourished.5 Complications include catheter failure, leakage from the insertion site, and peritoneal infection. If peritoneal dialysis fails, extracorporeal therapies (ie, hemodialysis) are often precluded because the patients are too small for the venous access required to deliver the therapy. In these cases, patients are often transitioned to palliative or comfort care because of the lack of a traditional KRT device for these small infants.
Over the past 5 years, several centers have reported using a modified ultrafiltration device to perform KRT in infants weighing 5 to 10 kg with good success.6–8 The benefit of this device, relative to traditional extracorporeal KRT devices, is that therapy can be performed using nontraditional, smaller access. Based on this pioneering work, we provided KRT to 7 newborns who were unable to receive traditional extracorporeal KRT because of their small size. The findings of this case series demonstrate that extracorporeal KRT is feasible in children weighing <2 kg and that the threshold for offering such therapy should be reevaluated. This case series was approved by the Stanford University institutional review board (#65240).
Kidney Replacement Therapy in Preterm Infants
Standard blood-based, extracorporeal KRT devices require intravenous access (7F hemodialysis catheter) that is too large for most patients weighing less than 2 to 3 kg because of their small vessel sizes. As such, traditional hemodialysis and continuous KRT are not possible. The dialytic approach using the modified ultrafiltration device described here can be performed despite the smaller vascular caliber in these infants. For these patients, we achieved access using 2 single-lumen lines: a 3F single lumen power-rated (3 mL/s) catheter was cut to the appropriate length and placed in the right femoral vein and a 4F single-lumen power-rated (5 mL/s) catheter was cut to the appropriate length and placed in the right internal jugular vein. All patients received intermittent hemofiltration (IHF) using a modified ultrafiltration device (Aquadex Smartflow, Nuwellis, Eden Prairie, MN). This device is traditionally used for continuous ultrafiltration therapy in patients with diuretic-resistant fluid overload.
Blood flow rates for this device range from 20 to 50 mL/min, which were easily achievable using the small single-lumen catheters. The priming volume of the circuit is 33 mL; blood priming was initially used in all patients and continued for as long as the circuit volume represented more than 10% of the patient’s circulating blood volume. Although some patients required inotropic support as part of their clinical care, routine use of vasoactives or volume infusions was not required to initiate therapy. Infants generally tolerated initiation well without alterations in their hemodynamics. The therapy was modified by running a physiologic replacement solution (PrismaSol, Baxter International, Deerfield, IL) into the prefilter tubing of the cartridge set via an existing y-connector. IHF was performed for 8 hours per day, targeting a standard 24-hour clearance of 25 mL/kg/h (75 mL/kg/h for the 8-hour session). Once euvolemia was restored, clearance was adjusted to deliver a minimum standard weekly Kt/V of 2.1 where Kt/V was calculated as follows: K = effluent rate (mL/h), t = time in hours, V = total body water. The electrolyte composition of the replacement fluid was modified according to individual patient needs. Therapy was initially provided without anticoagulation because of concerns around cerebral bleeding risks early in life. However, even with the intermittent nature of the therapy, blood flow rates became harder to achieve and circuit clotting was seen; eventually, heparin was used in all 7 patients and titrated to maintain activated clotting times of 120 to 150 seconds.
Results
Seven neonates (3 males and 4 females, Table 1) with ESKD were treated using IHF as KRT. The cause of ESKD was congenital anomalies of the kidney and urinary tract in 6 and twin-twin transfusion syndrome in the seventh. Three patients received IHF as their initial therapy and 4 received IHF as rescue therapy following failed peritoneal dialysis. Median gestational age and birth weight were 32 weeks (interquartile range [IQR], 32-35) and 1.58 kg (IQR, 1.41-2.01), respectively. At birth, all 7 weighed less than 2.5 kg and 4 weighed less than 2 kg. IHF was initiated at a median age of 16 days (IQR, 1.5-40) and a weight of 1.85 kg (IQR, 1.57-2.1). At initiation, median creatinine and percent of fluid overload were 2.8 mg/dL (IQR, 1.8-7.1) and 18% (IQR, 10-40), respectively. Median duration of IHF therapy was 96 days (IQR, 82-147).
Patient Demographics
| Patient 1 | Patient 2 | Patient 3 | Patient 4 | Patient 5 | Patient 6 | Patient 7 | |
|---|---|---|---|---|---|---|---|
| Gestational age, wk | 31 | 35 | 32 | 32 | 35 | 35 | 32 |
| Age at initiation, d | 73 | 16 | 55 | 25 | 2 | 1 | 1 |
| Sex | Male | Female | Male | Male | Female | Female | Female |
| Primary or rescue therapy | Rescue | Rescue | Rescue | Rescue | Primary | Primary | Primary |
| Birth weight, kg | 1.2 | 1.6 | 1.3 | 1.8 | 2.2 | 2.5 | 1.5 |
| Duration of therapy, d | 157 | 41 | 152 | 89 | 75 | 142 | 96 |
| Weight at initiation, kg | 1.5 | 1.9 | 1.6 | 2 | 2.2 | 2.5 | 1.5 |
| Fluid overload at initiation, % | 18 | 37 | 219 | 42 | 9 | 8 | 10 |
| BUN at initiation, mg/dL | 106 | 43 | 20 | 39 | 41 | 15 | 15 |
| BUN at 1 wk, mg/dL | 49 | 9 | 49 | 31 | 9 | 15 | 6 |
| Creatinine at start, mg/dL | 7.8 | 8.8 | 1.9 | 6.4 | 2.8 | 1.8 | 1.4 |
| Creatinine at 1 wk, mg/dL | 3.8 | 2.6 | 2.2 | 3.0 | 2.1 | 2.7 | 2.0 |
| Respiratory support at initiation | LFNC | SIMV | SIMV | RA | CPAP | SIMV | SIMV |
| Respiratory support at 1 wk | RA | RA | SIMV | RA | CPAP | CPAP | RA |
| Respiratory support at 30 d | RA | RA | CPAP | RA | RA | RA | RA |
| Patient 1 | Patient 2 | Patient 3 | Patient 4 | Patient 5 | Patient 6 | Patient 7 | |
|---|---|---|---|---|---|---|---|
| Gestational age, wk | 31 | 35 | 32 | 32 | 35 | 35 | 32 |
| Age at initiation, d | 73 | 16 | 55 | 25 | 2 | 1 | 1 |
| Sex | Male | Female | Male | Male | Female | Female | Female |
| Primary or rescue therapy | Rescue | Rescue | Rescue | Rescue | Primary | Primary | Primary |
| Birth weight, kg | 1.2 | 1.6 | 1.3 | 1.8 | 2.2 | 2.5 | 1.5 |
| Duration of therapy, d | 157 | 41 | 152 | 89 | 75 | 142 | 96 |
| Weight at initiation, kg | 1.5 | 1.9 | 1.6 | 2 | 2.2 | 2.5 | 1.5 |
| Fluid overload at initiation, % | 18 | 37 | 219 | 42 | 9 | 8 | 10 |
| BUN at initiation, mg/dL | 106 | 43 | 20 | 39 | 41 | 15 | 15 |
| BUN at 1 wk, mg/dL | 49 | 9 | 49 | 31 | 9 | 15 | 6 |
| Creatinine at start, mg/dL | 7.8 | 8.8 | 1.9 | 6.4 | 2.8 | 1.8 | 1.4 |
| Creatinine at 1 wk, mg/dL | 3.8 | 2.6 | 2.2 | 3.0 | 2.1 | 2.7 | 2.0 |
| Respiratory support at initiation | LFNC | SIMV | SIMV | RA | CPAP | SIMV | SIMV |
| Respiratory support at 1 wk | RA | RA | SIMV | RA | CPAP | CPAP | RA |
| Respiratory support at 30 d | RA | RA | CPAP | RA | RA | RA | RA |
BUN, blood urea nitrogen; CPAP, continuous positive airway pressure; LFNC, low-flow nasal canula; RA, room air; SIMV, synchronized intermittent mandatory ventilation.
Patients who received IHF as rescue therapy were 40 days old (IQR, 22.8-59.5) and weighed 1.73 kg (IQR, 1.58-1.89) at initiation. Median fluid overload at initiation was 40% (IQR, 32-86). In these patients (Fig 1, solid line), creatinine fell over the first 2 weeks of therapy and then plateaued, demonstrating excellent metabolic control. IHF effectively treated hypervolemia as the patients’ weights fell during the first 2 weeks of therapy preceding establishment of normal weight gain for the duration of therapy (Fig 2, solid line)
Serum creatinine values during the first month of life. Temporal trends in serum creatinine values (mg/dL) for our cohort are shown over the first 4 weeks of life. Patients who received intermittent hemofiltration (IHF) as rescue therapy (solid lines) experienced a rapid decline in serum creatinine during the first week of therapy and plateaued at levels appropriate for their underlying pathophysiology. Patients who received IHF as primary kidney replacement therapy (dashed lines) experienced stability in serum creatinine values for the entire month.
Serum creatinine values during the first month of life. Temporal trends in serum creatinine values (mg/dL) for our cohort are shown over the first 4 weeks of life. Patients who received intermittent hemofiltration (IHF) as rescue therapy (solid lines) experienced a rapid decline in serum creatinine during the first week of therapy and plateaued at levels appropriate for their underlying pathophysiology. Patients who received IHF as primary kidney replacement therapy (dashed lines) experienced stability in serum creatinine values for the entire month.
Weight during the first 5 months of life. Weights (kg) for our cohort are shown over the first 5 months of life, truncated at death or transfer to another center. Patients who received intermittent hemofiltration (IHF) as rescue therapy (solid lines) experienced a decline in weight during the first month of therapy as euvolemia was restored in the setting of preexisting hypervolemia; thereafter, these patients experienced physiologically appropriate weight gain. Patients who received IHF as primary kidney replacement therapy (dashed lines) experienced normal physiologic weight gain from birth through the first 5 months of life.
Weight during the first 5 months of life. Weights (kg) for our cohort are shown over the first 5 months of life, truncated at death or transfer to another center. Patients who received intermittent hemofiltration (IHF) as rescue therapy (solid lines) experienced a decline in weight during the first month of therapy as euvolemia was restored in the setting of preexisting hypervolemia; thereafter, these patients experienced physiologically appropriate weight gain. Patients who received IHF as primary kidney replacement therapy (dashed lines) experienced normal physiologic weight gain from birth through the first 5 months of life.
Patients who received IHF as primary therapy were 1 day old (IQR, 1-1.5) and weighed 2.2 kg (IQR, 1.87-2.35) at initiation. Median fluid overload at initiation was 9.4% (IQR, 8.9-10.1). In these patients (Fig 1, dashed line), creatinine remained stable throughout the first month of KRT, again demonstrating excellent metabolic control. Patients remained euvolemic during the first week of therapy before establishing normal weight gain for the duration of therapy (Fig 2, dashed line).
All 7 patients experienced excellent linear growth and appropriate increases in head circumference (Fig 3) over the course of their IHF therapy. Survival to hospital discharge was 71% (5/7) with 2 in-hospital deaths attributable to infection (1 viral and 1 bacterial), resulting in vasoplegic and/or septic shock. Of the surviving patients, 4 were transitioned to peritoneal dialysis as a bridge to kidney transplantation. Candidacy for peritoneal dialysis catheter placement was determined in consultation with the surgeon based on patient weight and abdominal size, adiposity, and nutritional status. The fifth patient was transitioned to hemodialysis and transferred to a facility closer to the family’s home. During IHF, 5/7 (71%) patients required a catheter exchange. No unexpected KRT-related complications occurred and no complications related to anticoagulation were seen.
Length and OFC during the first 5 months of life. Length (cm) and occipitofrontal circumference (OFC, cm) for our cohort are shown over the first 5 months of life, truncated at death or transfer to another center. Patients who received intermittent hemofiltration (IHF) as rescue therapy (solid lines) and patients who received IHF as primary kidney replacement therapy (dashed lines) experienced normal physiologic linear growth and an appropriate increase in OFC.
Length and OFC during the first 5 months of life. Length (cm) and occipitofrontal circumference (OFC, cm) for our cohort are shown over the first 5 months of life, truncated at death or transfer to another center. Patients who received intermittent hemofiltration (IHF) as rescue therapy (solid lines) and patients who received IHF as primary kidney replacement therapy (dashed lines) experienced normal physiologic linear growth and an appropriate increase in OFC.
Discussion
The decision to provide KRT to neonates can be fraught with ethical and technical challenges.9 Until recently, it was not possible to offer extracorporeal KRT to neonates under a certain weight because of restrictive vascular access and device options. A few centers performed pioneering work in the development of an alternative form of extracorporeal KRT using a modified ultrafiltration device.6,7 In patients weighing 5 to 10 kg, this approach offers technical advances over traditional therapies, including the ability to cannulate smaller caliber vessels and to use a reduced extracorporeal volume.
We have used this approach to deliver KRT in neonates for whom extracorporeal therapy was technically impossible because of an inability to achieve adequate access to use traditional devices. At our center, although a 7F dialysis catheter is required to provide hemodialysis or traditional blood-based continuous KRT, we were able to provide IHF using this modified ultrafiltration approach with significantly smaller catheters. These patients would not have been offered traditional extracorporeal KRT and would not have survived otherwise.
IHF rapidly corrected uremia and lowered serum urea and creatinine values to levels seen in neonates receiving other forms of chronic KRT. Additionally, in those patients who were fluid overloaded, we were able to restore euvolemia within the first 1 to 2 weeks of therapy. Once patients were on a stable IHF regimen, nutritional support and fluid intake could be liberalized to provide age-appropriate caloric delivery such that excellent weight gain, linear growth, and increasing head circumference were demonstrated. Although this therapy can be provided continuously, the ability to deliver intermittent KRT over an extended period may offer additional advantages, including opportunities for physical therapy, tummy time, and parental interaction. The use of IHF before peritoneal dialysis catheter placement allowed optimal healing of the catheter and tunnel site, potentially reducing complications of the therapy. Overall survival was good and consistent with that seen in children weighing <5 kg receiving continuous KRT.10
To summarize, extracorporeal KRT is technically feasible in preterm neonates between 1 and 2 kg and may be considered first-line therapy as a bridge to peritoneal dialysis for those who require it. Although our initial success is encouraging and the technique offers advantages over traditional approaches, further evaluation is required. Placement of these catheters and use of this device may be associated with short- and long-term complications, including infection and thrombosis; we do not yet know the long-term safety profile of this approach. It is also important to note that it requires specific, multidisciplinary expertise so that referral to a center experienced with this mode of KRT would be necessary. Although ethical considerations and patient complexity must be taken into account, technological advances have made it possible to offer extracorporeal KRT to neonates with AKI or ESKD who would not have previously been eligible based on weight alone.
Dr Sutherland participated in the care of the patients, conceptualized and designed the study, drafted the initial manuscript, and reviewed and revised the manuscript; Dr Davis, Mss Powell, Woo, and Tanaka, and Dr Josephs participated in the care of the patients and reviewed and revised the manuscript; Dr Wong participated in the care of the patients, conceptualized and designed the study, and reviewed and revised the manuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
FUNDING: No external funding.
CONFLICT OF INTEREST DISCLOSURES: Dr Sutherland has received a speaking honorarium from the manufacturer of the device used to dialyze the patients in this case series. The remaining authors have indicated they have no potential conflicts of interest to disclose.
Comments