Should We Screen Preterm Infants for Nephrocalcinosis? An Evidence-based Decision

Shahirose S Premji; Majeeda Kamaluddeen


Pediatr Health. 2010;4(1):24-35. 

In This Article

Appraisal & Judgments Related to Each of the WHO Criteria

What is the Natural History of Nephrocalcinosis in Premature Infants & is It an Important Health Problem?

Since nephrocalcinosis resolves in the majority of cases, the heterogeneity of the natural history of diseases introduces length bias, specifically overdiagnosis of nephrocalcinosis.[11] Hence, the risk of psychosocial harm with a false-positive screening test must be considered. As evident from anecdotal reports from one screening program, parents experienced unnecessary distress and anxiety.

Heterogeneity, both clinical (e.g., birth weight, gestational ages and age at study) and methodological (e.g., varying methods of assessing proximal tubular function, errors in sampling urine and study design), makes it challenging to conclude that neonatal nephrocalcinosis contributes to proximal tubular dysfunction. Controversy remains with regard to whether the renal dysfunction on follow-up reflects prematurity rather than morbidity associated with nephrocalcinosis.[17,29] With regard to hypercalciuria, differences in findings related to the calcium:creatinine ratio in preterm infants with and without neonatal nephrocalcinosis may relate to differences in measurement (e.g., spot urine,[29] random specimens of urine or 24-h urine collection[16]). In addition, urine extraction methods, frequency of stools and delayed collection of urine from a wet diaper may all contribute to errors in estimating urinary calcium.[38] There are numerous causes of hypercalciuria, including enteral and parenteral practices (e.g., excess calcium, inadequate phosphorus or excess vitamin D intake), and drugs (e.g., furosemide, methylxanthines and dexamethasone) to name a few.[16,39] Specific tests would be needed for confirming the diagnosis, and treatment would depend on the identifiable cause of hypercalciuria.[39] The literature does not provide empirical evidence that neonatal nephrocalcinosis further impedes growth of the kidneys.[25] Kist-van Holthe et al. concluded that "prematurity per se is associated with high blood pressure".[28] Consequently, there are significant gaps in our understanding of the natural history of nephrocalcinosis and the evidence is equivocal to determine whether nephrocalcinosis is an important health problem in the long-term.

How is Nephrocalcinosis Diagnosed in Preterm Infants?

Ultrasound has been described as a subjective method, which is "known for its remarkable operator dependence".[20] The use of ultrasound for diagnosing nephrocalcinosis is associated with variability in image quality as determined by ultrasound transducers and interobserver variability.[1,12] Image quality has the potential to increase inter- and intra-observer variability,[1,13] both of which have improved with the use of a 7.5-MHz ultrasound transducer when compared with a 5.0-MHz transducer.[1,12] In children at risk of nephrocalcinosis, a study using a hypothetical grading scale that was developed based on previous findings and experience to describe severity of nephrocalcinosis found good intra- and inter-observer agreement.[40] Despite improvements in technology, ultrasound has been criticized for its lack of specificity of the parenchymal changes; hence, it may not be the ideal test. Case reports of children suggest that computed tomography offers the advantage of recognizing very small differences in density and, consequently, may be better to detect very early stages of nephrocalcinosis and for assessing extent of disease.[41] However, no studies could be identified comparing sensitivity and specificity of ultrasound versus computed tomography in detecting nephrocalcinosis. The use of computed tomography in premature infants to diagnose nephrocalcinosis may not be feasible or cost-effective and would result in unnecessary exposure to radiation.

Are there Effective Evidence-based Treatments for Nephrocalcinosis?

Citrate therapy for prevention of nephrocalcinosis was not efficacious; hence, at present there is no evidence-based therapy for prevention of nephrocalcinosis. However, this conclusion is drawn from a small randomized, controlled trial.

Is there a Recognizable Latent or Early Symptomatic Stage & Does Treatment in this Phase Lead to Better Health Outcomes?

Treatment options can be explored when there is confirmation of disease or alternatively when a precursor is identified in someone at high risk of disease or suspected of disease. In premature infants, hypercalciuria, which is a risk factor for nephrocalcinosis, can be diagnosed by determining the calcium:creatinine ratio in a random urine sample.[6,39,42] The use of the calcium:creatinine ratio to screen for hypercalciuria is problematic as it has been recognized that virtually all premature infants begin life in a hypercalciuric state.[17] The calcium:creatinine ratio is high in infancy and decreases throughout childhood.[43] Studies examining the natural history of nephrocalcinosis seem to support this premise.[6,28,29] At present, there is a paucity of literature to determine whether a high calcium:creatinine ratio is a recognizable latent or early symptomatic stage.

Are there Screening Programs for Nephrocalcinosis & What has been Their Experience?

In 2003, a nephrocalcinosis screening programme was implemented in the Neonatal Intensive Care Unit of National Women's Health in Auckland, New Zealand, after a high incidence of nephrocalcinosis (83%) was identified in babies enrolled in a dexamethasone study [Bailie H, Kuschel CA, Wong W: Unpublished Abstract]. All infants born before 30 weeks' gestation were to be screened by renal ultrasound at 36 weeks' corrected gestational age or at discharge. Over a 2-year period (2003–2005), 70% of eligible babies were screened with the first follow-up scan performed at a median age of 14 months (mean 9–22 months). The incidence of nephrocalcinosis was 21%, in 86% of infants nephrocalcinosis resolved by 33 months without intervention, no affected infants had clinical manifestation and none were treated [Bailie H, Kuschel CA, Wong W: Unpublished Abstract]. Out-patient follow-up also included measuring urine calcium:creatinine ratios (n =10), which were initially elevated in three infants and normalized in two infants when repeated. Although degrees of nephrocalcinosis were evident on ultrasounds at 36 weeks' corrected gestational age, it was not feasible to predict resolution based on these findings [Kuschel C, May 2008, Pers. Comm.].

Anecdotal reports indicate that the screening programme was a source of distress and anxiety given that nephrocalcinosis may resolve spontaneously and as such may not be a problem [Kuschel C, May 2008, Pers. Comm.]. Professional Organizations such as the Canadian Pediatric Society and the American Academy of Pediatrics have no recommendations with regard to screening programs for nephrocalcinosis in premature infants.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.