Peritoneal Dialysis in Neonates and Infants

Introduction

Peritoneal dialysis (PD) has long been considered an effective treatment modality for neonates and infants with severe acute kidney injury (AKI) and is the dialysis modality of choice for patients in this age group with end-stage kidney disease (ESKD). Its popularity and success largely derive from its simplicity and effectiveness as a means of removing solute and fluid in even the smallest patients and because it is a well-tolerated procedure by most patients.

Peritoneal Dialysis as a Renal Replacement Modality for Acute Kidney Injury

There is limited literature documenting the etiology of AKI that results in the need for acute PD in this population. However, some generalizations can be made. In most cases, congenital malformations such as renal dysplasia or posterior urethral valves do not compromise kidney function so severely that dialysis is required in the newborn period. Instead, acquired renal disorders, usually related to perinatal asphyxia, hypoxia, sepsis, hypovolemia, and cardiac surgery, make up the majority of insults that mandate acute replacement therapy. Very rarely, acute vascular events such as renal artery or renal vein thrombosis are to blame, but these insults usually have to affect both kidneys to result in AKI requiring dialysis.

Despite the growing popularity of continuous renal replacement therapy (CRRT), survey results of pediatric nephrologists provide evidence that PD remains the predominant acute dialysis modality for children < 2 years of age. Access can be placed emergently at the bedside in patients who are too unstable to undergo a surgical procedure. Additionally, PD can typically be performed safely and effectively in patients with cardiovascular instability as a result of the inherent gradual and continuous provision of both ultrafiltration and solute clearance characteristics of the procedure. Recently, several studies have demonstrated improved outcomes associated with prophylactic placement of a PD catheter and early initiation of PD in infants undergoing cardiac surgery. One randomized prospective trial comparing PD versus furosemide in this population showed improved fluid management, less inotrope use, and a shorter duration of mechanical ventilation in the PD group. Finally, PD offers the additional advantage of not requiring systemic anticoagulation, in contrast to hemodialysis (HD) and CRRT.

There are rare absolute contradictions to PD in neonates, such as omphalocele, diaphragmatic hernia, or gastroschisis, all scenarios in which there is essentially the lack of a functional peritoneal cavity. Relative contraindications to PD include the presence of severe pulmonary disease since the increased intra-abdominal pressure associated with PD could further compromise pulmonary function, or a history of extensive abdominal surgery because of the possible presence of adhesions and a compromised peritoneal cavity. The presence of a vesicostomy or colostomy or the diagnosis of polycystic kidney disease is not a contraindication to PD, although the risk of peritonitis may be greater in this setting. With these issues in mind and given the lack of prospective trials comparing the outcomes of neonates and infants with AKI stratified by dialysis modality, the decision regarding dialysis modality selection always needs to factor in local resources and expertise along with the patient’s clinical status.

Peritoneal Dialysis Access in Acute Kidney Injury

The two most commonly placed accesses for acute PD are the percutaneously placed Cook catheter and the surgically placed Tenckhoff catheter. The Cook catheter offers the advantage of bedside placement by a nephrologist or intensivist via the Seldinger technique. Since only local anesthesia is required, it can be placed promptly, even in an unstable patient. However, its use has been associated with a high rate of complications such as leakage of dialysis fluid from the catheter entry site on the abdominal wall and catheter obstruction. Chadha et al., in a single-center retrospective study of infants and young children with AKI, found that by day 6 of dialysis, only 46% of Cook catheters were functioning without complications. In comparison, they found that over 90% of surgically placed Tenckhoff catheters were free of complications at the same time point. Thus the authors suggested that if acute dialysis is expected to be required for more than 5 days, a Tenckhoff catheter should either be placed initially, or elective replacement of the Cook catheter with a Tenckhoff catheter can be performed in a timely manner. Subsequently, the same center reported promising results with the use of a multipurpose percutaneous catheter (Cook Mac-Loc Multipurpose Drainage catheter) in a small cohort of infants with AKI who experienced a mean complication-free catheter survival of approximately 11 days. Finally, another recent small series described the successful use of a 16 G single-lumen central venous catheter placed as a PD catheter via the Seldinger technique in 5 infants with severe AKI after cardiac surgery.

The most important consideration for the successful placement and function of a Tenckhoff catheter in this population is the experience of the surgeon. This can be particularly problematic at centers caring for a small number of patients overall, where the need to provide dialysis to a very young infant may be a rare event. Because of the importance of the access and the desire for the outcome of placement to be complication free, the surgical placement should ideally be limited to only a few surgeons per center; on rare occasions, it may be preferable to refer the patient to another, more experienced center for access placement, in a manner similar to what has been recommended for vascular access.