Recurrent UTIs In Children: What Do We Know/Not Know?


Urinary tract infections (UTIs) are not uncommon in children, with some studies reporting an incidence of 2% in boys and 8% in girls under age 7. In the setting of vesicoureteral reflux (VUR), the risk of febrile UTI and recurrence of infections is heightened, which can result in long-term complications including renal scarring and chronic kidney disease.

One of the biggest challenges for pediatric nephrologists is deciding which patients are at risk for recurrence of UTIs and who may benefit from antimicrobial prophylaxis. Older studies have advocated for the use of VCUG to detect VUR in children after UTIs, as this is a known risk factor, but even children without reflux may be at risk for recurrence including those with other abnormal urinary tract anatomy or those who have bladder-bowel dysfunction. The PRIVENT trial randomized 576 children with a history of at least one prior UTI to receive trimethoprim–sulfamethoxazole or placebo and found an absolute risk reduction of 6% in the antibiotic group for recurrence of UTI during the 12-month study period. The two arms were comparable in baseline demographics and included children who had no VUR as well as varying degrees of VUR. This trial was much larger than older randomized studies from the 1970s, which also had fewer subjects with VUR.

So if antibiotic prophylaxis has a modest reduction in preventing UTIs in all-comers, how effective is it in children with known VUR? In the RIVUR study, 607 children with VUR identified after their first or second febrile or symptomatic UTI were randomized to trimethoprim–sulfamethoxazole or placebo and followed over a 2-year period. Antimicrobial prophylaxis decreased the risk of UTI recurrence (25.5% in treatment group vs. 37.4% in placebo group), with a number to treat of 8 children over the 2-year period to prevent one UTI. There was no significant difference between arms on level of adherence to providing the study medication. Prophylaxis was most effective in children with a febrile UTI as the index infection or who had concurrent bladder-bowel dysfunction. Risk of developing antimicrobial resistance was higher in the prophylaxis group (27.6% vs. 19.5%) though not statistically significant. The investigators also looked at whether antimicrobial prophylaxis could reduce the risk of worsening renal scarring or formation of new renal scars. Though they found no difference between groups, the study was not powered to detect statistical significance for this secondary outcome.

New evidence from these larger clinical trials support a role for antimicrobial prophylaxis in preventing recurrence of UTIs, particularly in high-risk patients such as those with VUR, and argues for a change in how to approach renal imaging in children with first-time UTIs. The 2011 AAP UTI clinical practice guidelines, based on a literature review performed at that time, argued that there was not enough trial data to support routine antimicrobial prophylaxis and thus prior guidance that recommended VCUG for first-time febrile UTI was not warranted; renal and bladder ultrasound was recommended, with VCUG in follow-up if ultrasound findings were concerning for VUR. This has led to some controversy, especially with the more recent RIVUR trial data supporting a role for antibiotic prophylaxis in VUR. A follow-up study looking at a historical cohort who had imaging prior to when these guidelines were published showed that the AAP guidelines would miss 56% of children in the cohort with grade 2 or higher VUR and all children with renal scarring who had a first-time febrile UTI. Despite these studies, the AAP reaffirmed its guidance in 2016, citing the benefits of cost reduction as well as decreasing patient discomfort and radiation exposure that goes along with VCUG. Though radionuclide cystogram (RNC), which has much lower radiation exposure, can be used to follow known VUR that has been diagnosed by VCUG, it doesn't provide the same level of anatomic detail for grading.

Additional large-scale trials are needed to determine whether antimicrobial prophylaxis, by way of decreasing UTI recurrence, can lower the risk of new renal scarring in children with VUR.


Neonatal Hyperammonemia: Dialyze, Scavenge, or Both?


We recently had a string of consults for neonatal hyperammonemia, each managed slightly differently. Hyperammonemia in neonates is a life-threatening emergency that can lead to severe long-term neurologic sequelae if not managed quickly. Though rare, hyperammonemia in the neonatal period is commonly due to inborn errors of metabolism, particularly urea cycle defects. The urea cycle is a multistep process responsible for removal of nitrogen from the bloodstream, and a defect in one of the enzymes can lead to buildup of ammonia which can cause brain damage or death.

Long-term treatment requires the use of nitrogen scavengers, such as sodium benzoate, sodium phenylacetate, and more recently carglumic acid in patients who have specific N-acetylglutamate synthase deficiency as a cause of their defect. In the acute presentation of neonatal hyperammonemia, dialysis remains the standard of care. 

How best to dialyze a neonate with hyperammonemia? The ultimate goal is rapid reduction of ammonia from the bloodstream and sustained clearance once the ammonia level has been brought down. Over the past few decades, HD techniques (either intermittent HD or CRRT) have been preferred over PD due to more rapid ammonia clearance. One approach is to use intermittent HD over a prolonged period (longer than a typical IHD session but not running over 24 hours like CRRT) to bring the ammonia level down quickly and then switch to CRRT to keep clearing ammonia as it is generated. One of the challenges with performing intermittent HD in hyperammonemic neonates, however, is their small body size and large blood flow rates you would need to use to clear ammonia effectively, which increases the risk of hemodynamic instability. Unlike intermittent HD, clearance in CRRT is dependent on dialysate flow rate and thus high-dose CRRT using dialysate flow rates above that which would typically be used in AKI has become more commonplace. This can prevent having to switch dialysis machines during treatment and allows for easy titration of the dialysis flow rate in response to serial measurement of serum ammonia levels.

Another controversial subject is whether or not you treat with nitrogen scavengers, such as sodium benzoate or carglumic acid, during the acute hyperammonemia phase or wait until the ammonia level has been brought down. This is always discussed with the Metabolism team. Though starting a nitrogen scavenger upfront may prevent a significant rebound from ongoing ammoniagenesis, these medications are dialyzable and thus higher IV rates may be required to be effective.

Each institution has its own approach. A good review on the topic and the controversies around dialysis in this setting can be found here. How do you approach hyperammonemia in neonates?

NephMadness 2018 Has Begun!

Just a couple hours ago, NephMadness 2018 was unveiled and regions have been announced. It should be no surprise that I will be rooting for the Pediatric Nephrology region (follow #PedsRegion on Twitter). Learn more about NephMadness here, including all the teams competing for top bragging rights, then submit your bracket.

For some NephMadness fun I created a crossword puzzle. Test your knowledge on pediatric kidney disease!

The Potassium Police


You get a call at 1:00am from the ICU resident about a neonate with a critical lab value. The serum potassium is 7 mmol/L. “We repeated the level as it was hemolyzed, but now the lab says it isn’t hemolyzed and level is 6.3. We’re getting an EKG and giving calcium gluconate.”

For a pediatric nephrology fellow, this is the kind of consult we often do in our sleep.

This is one of the most common consults we get, when an infant has a blood draw and is found to have hyperkalemia with no other explanation such as AKI, CKD, endocrine disruption (always check the newborn screen!), or medication side effect (ACE inhibitors, NSAIDs, beta-blockers, and trimethoprim-sulfamethoxazole are common culprits we see). Often if the serum level is elevated we’ll check a whole blood potassium level as this is less likely to have hemolysis during processing and is less reflective of potassium released by platelets during the clotting process in serum sample preparation.

Potassium homeostasis in infants (and children in general) is different than in adults. Unlike in adults, who must balance potassium intake and excretion, children need to be in positive potassium balance to ensure adequate somatic growth. There is a developmental immaturity of the kidneys with an end result of a decreased potassium filtration fraction and limitation on the ability to excrete an acute potassium load. Some research suggests that renal potassium handling in children may not reach that of adults until they reach early adolescence.

The distal nephron is the major contributor to the physiologic hyperkalemia seen in infants. This segment of the nephron is relatively resistant to mineralocorticoids, and infants have physiologically higher levels of plasma renin and aldosterone levels as a consequence. In addition, the distal nephron has decreased expression of two key apical membrane channels, the ROMK and maxi-K (or BK) channels, which permit potassium efflux down its concentration gradient from the intracellular space to the luminal fluid. Whereas ROMK is important for basal potassium excretion, BK activity is augmented by increased urinary flow through the distal tubule. There is also low expression of the epithelial sodium channel (ENaC), which is localized to the principal cells of the collecting duct. This limits potassium efflux in response to the generation of a negative luminal charge when sodium passes through ENaC. Finally, throughout the whole nephron, infants have decreased Na-K ATPase activity which diminishes the concentration gradient to allow for effective transmembrane sodium and potassium exchange.

Developmental differences in neonatal and infant renal potassium handling compared to the adult nephron. From  Pediatr Nephrol (2017) 32:2037–2049 .

Developmental differences in neonatal and infant renal potassium handling compared to the adult nephron. From Pediatr Nephrol (2017) 32:2037–2049.

So how do we overcome this physiologic immaturity of the kidneys and manage hyperkalemia in these infants? Give some salt! If you were to check the urinary sodium concentration in these infants, a majority of the time the level will be low or undetectable. Breast milk or formula that infants consume has low sodium content and exacerbates the problem. A small amount of sodium chloride supplementation is often enough to increase Na-K ATPase activity as well as activate the ENaC channels on the principal cells of the collecting duct to drive potassium into the final urine.

Based on this developmental immaturity of renal potassium handling, certain infants have a higher risk for hyperkalemia and may warrant closer monitoring:

  • Preterm infants
  • Infants on diuretics for chronic lung or congenital heart disease (though initially they may become hypokalemic with therapy, they can become hyperkalemic as their total body sodium is depleted)

  • Infants with short gut syndrome/malabsorption (decreased ability to absorb sodium from the GI tract, often more difficult to manage with sodium chloride alone)

In an emergent situation, when an infant has hyperkalemia with EKG changes with or without hemodynamic instability, emergent measures are needed to protect the heart (calcium gluconate) and rapidly promote transcellular shift such as insulin/dextrose, albuterol, and sodium bicarbonate. However the ultimate goal is to remove potassium from the body, which can be through the urine (e.g. loop diuretics) or the GI tract (e.g. sodium polystyrene sulfonate). Despite its usefulness in these situations sodium polystyrene sulfonate is avoided by some institutions, especially in neonates, because of the small but real risk of intestinal necrosis. For infants with stable hyperkalemia, pre-treating the formula or breast milk with sodium polystyrene sulfonate powder is preferred and effective. Though patiromer is now being used for hyperkalemia in adults, the absence of studies in infants and children makes it not quite ready for primetime.

*Note: The above post does not constitute medical advice. It is commentary on how I approach hyperkalemia in neonates and infants and is provided for educational purposes only.

Simple Things Done Well

At the beginning of each term I chat with our new residents and fellows to find out what excites them about paediatric nephrology, aiming to tailor their project and exposure to their interests. Many of our junior team are planning to do general paediatrics (inevitably part-time, in the country, with a special interest in neonatology) and the fellows express a fascination with transplant, haemodialysis, or genetics. What they are never interested in though, is chronic kidney disease (CKD).

The management of CKD is an exercise in simplicity and tedium, following the same straightforward checklist for each patient. There are no accolades for good CKD management. No thanks. No awards. If managed well then their GFR declines at a slower rate and they feel better. But, hey, we’re paediatric nephrologists, right? We don’t need thanks, we just want our patients to grow up big and strong, and we don’t ever want them to feel sick.

After reading The Checklist Manifesto I became even more convinced that the consistent application of lists could improve patient care, particularly in this neglected area of paediatric nephrology. With this in mind I created a checklist which has migrated to a smartphrase in my hospital EMR, to make sure I optimise medical management of CKD for every patient, every time.

Iron – Check Hb and Iron studies. Kidney tissue needs oxygen to work effectively. The prevention of anaemia decreases CKD progression, decreases LVH, decreases sensitizing transfusions and increases quality of life. Supplement with oral iron early on and consider IV as necessary. Consider erythropoietin stimulating agents in stage 4 CKD when the kidney produces less erythropoetin.

Vitamin D – Most kids have low levels of 25-hydroxyvitamin D and I aim to keep levels >75nmol/L. Post hoc analysis of the ESCAPE trial showed 5-year renal survival was 75% in patients with baseline 25-hydroxyvitamin D ≥50 nmol/L and 50% in those with lower 25-hydroxyvitamin D levels (P<0.001). Vitamin D may promote renal protection by suppressing renin transcription through crosstalk between the renin-angiotensin-aldosterone and vitamin-FGF23-Klotho pathways. Supplement with an age appropriate dose (most get 1000 IU/day) and consider stress dosing with caution (e.g. 100,000 IU) in clinic to get the levels up and use regular dosing for maintenance.

Phosphate – Aim to keep levels around the 50th percentile for age, remembering that levels are lowest in the morning so hovering above the normal range at that stage is not good news. Remember that vascular calcification starts in early CKD and is a major cause of morbidity and mortality for our patients in adulthood. Don’t wait until the chemistry is abnormal, talk to families about limiting phosphate containing foods in early CKD, refer to your renal dietician, and consider phosphate binders if levels are sneaking up despite good dietary control. Considering a vegetarian diet is also a sensible move perhaps because non-animal phosphate appears to be less avidly absorbed.

PTH – Aim to keep this in the normal range. Around CKD stage 3 the kidneys are less efficient at converting 25-hydroxyvitamin D to the active 1,25-dihydroxyvitamin D, so give the smallest supplement you can to maintain normal PTH and calcium levels.

Blood Pressure – Normal blood pressure reduces glomerular pressure and long standing damage. Initial non-pharmacological measures such as weight reduction, exercise and salt restriction are good for the kidneys on many levels. But if hypertension persists then controlling BP to the 50th percentile (if tolerated) is the goal, since the ESCAPE trial showed a significant reduction in GFR decline in those controlled to the 50th percentile.  Following the comprehensive AAP guidelines is your best bet and I tend to choose and ACE inhibitor if proteinuria is also present.

ProteinuriaControlling proteinuria should reduce damage to the glomerulus. An increase of creatinine of 20% with the use of an ACE inhibitor is acceptable if the proteinuria is controlled. Warn parents in advance about this and remember to discuss the potential for teratogenicity in post-pubertal girls using an ACE inhibitor.

Acid-Base – Aim to maintain bicarbonate levels above 22 mEq/L with oral supplementation to prevent bone buffering, growth failure and progression of CKD.

Fluid Management – In early CKD many of our patients (particularly those with obstructive uropathy or nephronophthisis) can have a relative urinary concentrating defect. This results in polyuria and polydipsia which often isn’t noticed unless we ask about heavy diapers, frequent urination, bedwetting and increased thirst in the morning. Generally this can be easily managed by parents allowing free access to water but it’s worth discussing why a child might be grumpy and thirsty in the morning and the importance of seeking early medical attention for a vomiting illness to prevent AKI.

The above list applies to my patients in early CKD, and represents my opinion and clinical practice. I’d be delighted to hear from readers of this blog as to their views, biases and approaches to the management of patients in early CKD.

The Shape Of Urine: My Path To Nephrology

     As a first assignment in the NSMC 2018 internship, we have been tasked with writing a blog post about any topic we choose. I decided to reflect on my path to pursuing a career in pediatric nephrology.


     Not all medical students or residents find a career in nephrology the same way. Mine began with a fish tank. I was an amateur tropical fish enthusiast as a teenager, and kept a freshwater fish tank displayed in our family’s basement. It must be kismet that I now read journal articles where the same fish I used to keep, Danio rerio, is used to study renal development and genetics. Anyway, I quickly understood the importance of the aquarium’s filter in removing waste and toxins to keep my fish healthy. One summer after returning from our beach vacation, I found the filter not running – likely broken for days - and my zebrafish and cichlids were floating lifeless at the top of the tank. Maybe if I took better care of the tank before the trip, the filter would not have failed.

     I went through college and my pre-clinical years of medical school learning the functions of all the other organ systems, but I was especially drawn to renal physiology. Sure it is complex, but after reading through a copy of Burton Rose’s tome Clinical Physiology of Acid-Base and Electrolyte Disorders I saw its elegance and logic. Like the aquarium filter, I understood how the kidneys too must be cared for to perform their many functions to remove toxins and tightly regulate acid-base, fluid, and electrolyte balance in our bodies. How the kidneys work to attain a fine balance of the milieu intérieur fascinated me and I wanted to learn more.

     Building on what I learned in the lecture halls, I discovered in residency that the kidneys do not work in a closed system.  One of my mentors, in between pearls of nephrology wisdom, would always say, “The kidney is the soul of the body.” I gained a true sense of what he meant as I spent more time on the wards. I took care of children with diverse presentations of disease where the kidneys were at their core, such as dehydration and severe hypernatremia in a baby with nephrogenic diabetes insipidus, failure to thrive in a child discovered to have bilateral renal dysplasia, and hypertensive encephalopathy in an adolescent with lupus nephritis.  To understand pediatric kidney disease requires the clinical acumen to recognize how kidney dysfunction can affect the child as a whole.


     Shaped by these experiences, I went into nephrology fellowship after pediatric residency and have loved every moment of it. My interest in social media as a tool for learning and teaching nephrology topics has run parallel with my formal training experience. By participating in the NSMC internship, I want to develop the skills necessary to share my appreciation for nephrology with other medical trainees and convince them to pursue this rewarding field as well.


I'm An Intern Again! The Nephrology Social Media Collective

This year I'm participating as one of the 2018 interns in the Nephrology Social Media Collective (NSMC) internship. Now in it's 4th year running, the goal of the internship is to train physicians in how to use Twitter and other social media platforms to communicate effectively online and develop your personal online identity. This year there are a record-breaking 25 interns who will be participating in all things related to nephrology in social media, including NephJC (, NephMadness (, Renal Fellow Network ( and others. Nimra Sarfaraz, one of this year's interns, got a great shot of the first conference call last night.


In addition to posting on Ins & Outs, I will be cross-posting content that I will be contributing to the NSMC. Stay tuned!

Measure Twice, Inflate Once

This is a real case, with patient information altered to maintain anonymity.

I recently saw an 10-year-old boy in clinic for evaluation of hypertension. He has type 1 diabetes that was diagnosed at age 2, which has been poorly controlled and requiring multiple admissions for hyperglycemia, a couple times associated with DKA. His BPs at prior outpatient clinic visits had been normal, but during his most recent admission he had automated systolic BPs up to the 140s-150s mmHg, but with manual auscultation they were in the 130s mmHg range. He had no hypertensive symptoms. He is tall for age (>97th percentile) but also obese (BMI 98th percentile). He had normal renal function (serum creatinine 0.5 mg/dL). A renal ultrasound was obtained while he was inpatient, which was normal.

When I saw the patient in clinic, he had a normal manual blood pressure of 109/66 and normal four extremity BPs. Due to his obesity he required an adult-sized BP cuff. Based on the new 2017 AAP Pediatric Hypertension guidelines with updated BP tables, his 90th percentile was 113/75, 95th percentile was 118/78, and 95th + 12 mmHg was 130/90.

Except for his body habitus and type 1 diabetes, there was nothing on history or physical exam that was suggestive of a secondary cause of hypertension. He had no proteinuria, microalbuminuria (early marker of diabetic kidney disease, which in this poorly-controlled patient is a real possibility), or signs of nephritis on urinalysis. Being ill in the hospital can increase your BP, but by 30-40 mmHg is unusual. Upon further questioning, the patient's father states, "Oh yeah I forgot, they used the blue cuff in the hospital," referring to the small adult cuff that would have barely fit around the patient's arm.


In children, BP measuring technique is critical. Using a BP cuff that is the wrong size for the patient is one of the more common reasons for pseudohypertension. Cuffs that are too large will underestimate the true BP and those that are too small with overestimate it. For assessing proper cuff size, the length of the BP cuff bladder should cover about 80% of the patient's arm circumference and the width should cover about 40% of the arm circumference. All blood pressure measurements should be obtained in an upper extremity, preferably in the right arm. Oscillometric devices can be used to screen for hypertension, but any elevated BPs should be confirmed with manual auscultation. And did you know, despite what Google images shows, the bell of the stethoscope should be used for auscultation?

For proper measurement, which is hard to do in kids (and why it's so important to have multiple BP measurements at different visits before confirming hypertension), the patient should ideally be sitting up straight, quiet, with feet on the ground. The arm should be supported by the examiner, with the upper arm at the level of the heart. BP variability simply from differences in body or arm position are surprising, and I included a table below of some examples (not comprehensive) of how improper technique can affect BP measurement.


Should We Vaccinate Our Kids (After Renal Transplant)?


Working at a children's hospital with a busy renal transplant program, our goal is to get children caught up with any vaccinations they are eligible to receive before they receive a donor kidney. Children with CKD should be vaccinated according to the same CDC schedule as children without kidney disease. However, they may require boosters prior to renal transplantation for vaccines that don't induce as optimal an immunogenic response in CKD, particularly pneumococcus and hepatitis B.

What about after renal transplantation? Sometimes children are unable to complete a vaccine series due to age or because a new vaccine is brought to market after they receive their transplant, such as the HPV quadrivalent vaccine.  Live vaccines, such as MMRV and rotavirus, are generally contraindicated in immunocompromised patients due to risk of reactivation and life-threatening disease. But what about the safety and efficacy of inactivated vaccines?

Many recent studies on the efficacy and safety of post-transplant vaccination have focused on the inactivated influenza vaccine, since it is given annually and allows for easier monitoring. A survey of United Network for Organ Sharing (UNOS)-certified kidney transplant centers demonstrated an increase in influenza vaccination rates from 84% of responding centers in 1999 to 95% in 2009. The minority of transplant centers that continue to not routinely vaccinate post-transplant commonly cite a lack of immunogenicity and concerns of safety, particularly with the risk of causing acute rejection by inducing de novo production of anti-HLA antibodies.

Immunogenicity may be lower with certain vaccines, such as the influenzaHPV, or meningococcal vaccines, if given early after renal transplantation (usually within the first 3-6 months, especially if induction therapy is provided). In a Midwest Pediatric Nephrology Consortium study, the rate of seroconversion with the influenza vaccine was not significantly different between renal transplant recipients and controls. There was no difference in seroconversion based on type of immunosuppressive regimen (steroid-free vs. steroid-based), but antibody titers 8 weeks post-vaccination in both renal transplant groups were significantly lower than the control group. Furthermore, antibody titers in response to other vaccinations given after transplantation, such as that seen with pneumococcus, may decline more rapidly than in controls. Though immunogenicity of vaccines given after transplant may be suboptimal, the risk of serious morbidity and mortality is much higher in immunocompromised patients from the diseases these vaccines prevent.

Though there is concern that vaccination post-transplant can lead to acute rejection, there is limited evidence to support this. A Swiss study assessed the risk of de novo anti-HLA antibody production in two cohorts of adult kidney transplant recipients after receiving a triple injection of two doses of the H1N1 vaccine and the seasonal influenza vaccine. 17.3% of subjects in the first cohort (16 of 92) and 11.9% of subjects in the second cohort (7 of 59) had new anti-HLA antibodies detected 6 weeks after completion of the immunization series. In 6 month follow-up of 20 subjects from both cohorts who had developed anti-HLA antibodies, 13 developed donor-specific antibodies (DSA) and two had biopsy-proven episodes of kidney function decline that could have been attributed to anti-HLA antibodies (one had thrombotic microangiopathy, one had antibody-mediated rejection). Another study looking at H1N1 vaccination found only a 24-28% increase in seroprotection after vaccination between two cohorts, and while there were no clinical rejection episodes nearly 12% of the first cohort developed new anti-HLA antibodies. Other studies have shown opposing data, and current clinical practice guidelines on vaccination in solid organ transplant recipients advise that there is no proven link between vaccination and the risk of acute rejection.

How does your transplant center approach vaccination after renal transplantation?