Traditionally, secondary hypertension has been considered the most common form of hypertension in childhood. However, recent trends indicate a shift in this paradigm [10,11,12,13,14,15,16]. In our retrospective study of 70 children with hypertension (mean age at presentation: 12.8 ± 4.7 years), 41.4% were found to have a secondary cause of hypertension. This finding aligns with recent literature, which indicates that essential hypertension is now the predominant form of pediatric hypertension, especially in patients over 6 years of age. This trend is largely attributed to the increasing prevalence of obesity and metabolic syndrome in children [10,11,12,13,14,15,16].

In fact, as age increased, the prevalence of secondary hypertension decreased (Supplementary Fig. 3). However, we want to highlight the high rate of secondary hypertension among 17-year-old patients. At this age, all cases of hypertension were related to CAKUT, likely because, with increasing age and after pubertal maturation, the congenital reduced nephron mass in patients with CAKUT becomes clinically evident, contributing to the development of hypertension.

Accurately identifying patients with secondary hypertension is crucial to effectively treating any underlying illnesses [1, 2]. Current guidelines recommend a case-by-case approach, considering factors such as age, medical history, clinical presentation, or treatment resistance [1, 2]. Interestingly, only 30 out of 70 patients in our cohort exhibited symptomatic hypertension, with headache being the most common symptom. Among these, only six symptomatic patients were found to have a secondary cause of hypertension, suggesting that the presence of symptoms alone is not a reliable indicator of secondary hypertension. Nevertheless, when symptoms are atypical or differ from headaches, the likelihood of secondary hypertension may increase.

To minimize diagnostic errors resulting from individualized and variable approaches, we implemented a standardized, comprehensive screening protocol in our clinical practice for all hypertensive patients. This approach aimed to ensure the identification of all secondary forms of hypertension. Notably, we also investigated potential secondary causes in patients whose characteristics, such as obesity, might typically suggest primary hypertension. Interestingly, some of these patients with obesity were ultimately diagnosed with secondary hypertension. This finding is consistent with the study of Kapur et al., which highlights that factors like obesity and the severity of hypertension should not preclude evaluation for secondary hypertension [17].

We analyzed our comprehensive screening for hypertensive children by differentiating between abnormal findings and contributory results and assessing the diagnostic utility of each test. Additionally, we conducted a cost analysis of the diagnostic work-up. Ding et al. [18] performed a similar study but without evaluating costs. They identified high-yield investigations, including kidney US, lipid profile for overweight or obese children, and echocardiograms to assess target organ damage. However, their study did not apply a uniform protocol of investigations to all participants [18]. As a result, the diagnostic yield was influenced by pre-test probabilities [18]. Similarly, Wiesen et al. [16] retrospectively evaluated the effectiveness of the protocol recommended by the Fourth Report [6] in patients with mild-to-moderate hypertension. Their findings indicated that routine urinalysis and serum biochemical tests were not particularly useful for diagnosing mild-to-moderate hypertension. The fasting lipid profile was the only blood test that consistently produced significant abnormal results. Additionally, they recommended the use of kidney US in cases with abnormal ABPM results or a strong clinical suspicion of renovascular hypertension or kidney parenchymal disease. However, this study had limitations including the lack of uniform examinations and the absence of a consistent, comprehensive diagnostic work-up.

In our cohort, kidney parenchymal or renovascular impairment emerged as the most common cause of secondary hypertension, consistent with findings from other studies [12, 13, 18]. Consequently, kidney US was the most contributory investigation in both the extended and short protocols. Abnormal creatinine levels were associated with CAKUT-related CKD. While urinalysis proved useful in evaluating hypertension, the ACR was less informative and primarily abnormal only in children with already abnormal urinalysis results. Urinalysis, however, demonstrates good sensitivity as a screening tool for macroalbuminuria (ACR > 300 mg/g), though its specificity is limited [19, 20]. Therefore, to detect also the cases of microalbuminuria (ACR between 30 and 299 mg/g), the ACR should be preferred [19, 20].

Hyponatremia was observed in one patient with renal artery stenosis, alongside hypokalemia, metabolic alkalosis, and abnormal findings on renal artery Doppler US. Similarly, the other patient in our cohort with renal artery stenosis exhibited hypokalemia, metabolic alkalosis, and abnormal Doppler US findings (Fig. 2). Doppler results were confirmed using CT/MRI angiographies, which also revealed abnormalities. The biochemical profiles of patients with renal artery stenosis in our study are consistent with the literature, as renal artery stenosis is known to be associated with metabolic alkalosis, hyperkalemia, and hyponatremia [21,22,23]. As described in the literature [24], hypokalemia was also identified in a patient with Cushing syndrome associated with the Carney complex, accompanied by abnormal cortisol, ACTH, and DST (Fig. 2). Abnormal lipid profiles were exclusively observed in children with overweight or obesity.

None of our patients showed elevated urinary levels of metanephrines, normetanephrines, adrenaline, noradrenaline, dopamine, vanillylmandelic acid, hydroxy-indoleacetic acid, or homovanillic acid that led to a diagnosis of pheochromocytoma as a secondary cause of hypertension. This is probably because pheochromocytoma or paraganglioma is more common in young adults, particularly between the ages of 30 and 40 [25]. However, we observed false-positive results in 28.6% of cases for catecholamines and in 12.8% for homovanillic acid.

It is crucial to emphasize that not all abnormal findings result in definitive diagnoses. Therefore, careful consideration is required to determine which patients should undergo further investigations. Universal testing may introduce confounding factors and create economic challenges.

While our extensive screening effectively diagnosed all secondary causes of hypertension, it incurred unjustifiable costs (€ 253.08 per patient) and produced a notable proportion of false-positive results. To address these limitations, we developed and tested a streamlined diagnostic protocol aimed at identifying all causes of secondary hypertension without missing diagnoses, while achieving a better cost–benefit ratio. We also compared this short diagnostic work-up with the guidelines of AAP and ESH (Table 3) [1, 2].

When comparing our extended and short diagnostic work-up, we found that the short protocol successfully identified the causes of hypertension or raised suspicion of a secondary hypertension in all patients identified by the extended protocol. Moreover, implementing the short diagnostic work-up resulted in a 64.3% cost saving in reaching the final diagnosis. The findings from the comparison of our short diagnostic work-up with the AAP and ESH guidelines [1, 2] are particularly noteworthy.

Both AAP and ESH protocols are less expensive than our short diagnostic work-up, with the AAP being the least costly (Table 3). However, the AAP protocol failed to identify several cases of secondary hypertension. This was primarily because it includes kidney US only for patients under 6 years old or for those with abnormal urinalysis or kidney function. In contrast, the ESH protocol, when applied to our cohort, successfully identified all cases of secondary hypertension. Both protocols included tests with a 0% rate of contributory findings, highlighting opportunities to further improve cost-effectiveness.

A detailed comparison between our short protocol and the ESH protocol [2] reveals that the main factor contributing to the cost difference is the routine use of Doppler US in our protocol. In the two patients with renal artery stenosis in our cohort, both had electrolyte abnormalities, and one exhibited kidney hypoplasia on US. As a result, the ESH protocol would have identified these patients as candidates for further investigation.

We acknowledge the limitations of Doppler US in diagnosing renal artery stenosis in children, including its susceptibility to false-negative and false-positive results. Nonetheless, we included Doppler US for all patients because it is a widely available, non-invasive imaging modality that avoids the risks associated with radiation and contrast agents [26]. Although its diagnostic accuracy is imperfect, Doppler US can identify patients with a higher suspicion of renal artery stenosis, guiding the need for confirmatory testing using more definitive modalities [26].

However, relying solely on Doppler US for diagnosis is insufficient. Based on the performance of the ESH protocol in our cohort, we propose that Doppler US should not be used routinely as a screening tool but reserved for cases with clinical suspicion of renal artery stenosis (e.g., electrolyte abnormalities, renal hypoplasia, poorly controlled hypertension).

Both AAP and ESH protocols also include tests such as urea, chloride, calcium, phosphorus, magnesium, and lipid profile [1, 2]. Additionally, the ESH protocol incorporates uric acid [2]. However, none of these tests showed contributory findings in our cohort and may not be strictly necessary. The ESH protocol also includes urine PCR and ACR [2], which in our cohort were abnormal only in patients with pre-existing abnormal urinalysis. These could potentially be omitted as routine screening tools, favoring urinalysis alone, as recommended by the AAP guidelines [1]. However, it is important to note that cases of isolated microalbuminuria (ACR between 30 and 299 mg/g) could be missed if only urinalysis is performed [19, 20].

We hypothesize that combining elements of our short diagnostic work-up with the AAP and ESH protocols could yield a cost-effective approach to screening for secondary causes of hypertension in children. This streamlined protocol would include serum creatinine, fasting glucose, sodium, potassium, urinalysis, and kidney US, with an estimated cost of € 44.30 per patient. Future studies and expert consensus are necessary before implementing this protocol in routine clinical practice (Table 4).

This study has several limitations, including its retrospective, single-center design, which requires validation in larger populations. However, single-center enrollment allowed for consistent clinical and radiological management, as well as standardized laboratory procedures. Furthermore, our center’s practice of extensive initial screening ensured that diagnostic yield was not influenced by the pre-test probabilities [18]. Selection bias is another limitation, as our university center specializes in managing children with CAKUT. Additionally, we often care for patients with complex endocrinopathies and diabetes, which necessitate hypertension evaluation. This likely accounts for the high rate of secondary causes observed in our study. Nevertheless, we believe this population reinforces the findings of our study by demonstrating the effectiveness of different protocols in screening for secondary hypertension in a high-risk cohort. However, our population is poorly representative of patients aged < 5 years, and additional data are needed in this age range. Finally, conditions such as pheochromocytoma or paraganglioma may be missed with the short protocol. Therefore, in cases presenting with typical symptoms (e.g., flushing and hypertensive peaks) or when hypertension is resistant to multiple drugs, a targeted diagnostic work-up should be initiated.

In conclusion, our study reinforces the recent trend indicating that primary hypertension is more prevalent than secondary hypertension in children, with kidney parenchymal disease being the most common cause of secondary hypertension. The short diagnostic work-up, costing € 90.25 per patient, successfully identified all cases of secondary hypertension detected by the extended protocol, achieving a 64.3% cost savings compared to the extended approach.

The most cost-effective strategy for screening secondary hypertension, however, appears to be the ESH protocol, with a cost of € 60.30 per patient (Table 3). By integrating elements of our short diagnostic work-up with the ESH and AAP protocols, focusing only on contributory tests, the cost could be further reduced to € 44.30 per patient without missing any diagnoses of secondary hypertension (Table 4).

When evaluating investigations in hypertensive children, it is crucial to emphasize that an abnormal result does not necessarily equate to a contributory finding. Thus, this study supports the AAP and ESH recommendations [1, 2], offering guidance on adopting a more cost-effective, simplified, and focused approach to managing children with elevated blood pressure. Additional testing can be tailored based on the context and the results of initial evaluations.