Conservative management of perinatal adrenal masses in a single institution retrospective study

Abstract

Purpose

Prenatal ultrasound screening has increased detection of adrenal masses during the fetal and neonatal periods, creating management dilemmas between immediate surgical intervention and observation. This study evaluates the outcomes of selective conservative management in perinatal adrenal masses.

Methods

We conducted a single-institution retrospective case series of nine patients with perinatal adrenal masses diagnosed at our institution between 2015 and 2025. Clinical presentation, imaging characteristics, biochemical markers, treatment decisions, and outcomes were analyzed. The cohort included five males and four females, with masses detected prenatally in five cases. All patients were initially managed conservatively with serial imaging surveillance to assess lesion volume over time.

Results

Neuroblastoma was confirmed in four patients; other diagnoses included four benign lesions and one adrenocortical carcinoma. Given the small sample size of this case series, the following proportions should be interpreted with caution and are not intended to reflect population-level outcomes. Spontaneous regression occurred in six cases, two remained stable, and one increased in size. Four patients underwent surgical intervention based on growth patterns or radiological suspicion of malignancy. Median follow-up was 42 months (range: 6 months to 10 years), with no mortality recorded.

Conclusion

Among the nine case, we observed spontaneous regression in six of nine patients. Temporal evolution of lesion volume proved more informative for clinical decision-making than initial imaging characteristics. The adrenocortical carcinoma case illustrates that radiological trend alone does not always determine management, and that clinical context—including genetic predisposition—must guide decisions. Although our small cohort precludes generalizable conclusions, this institutional experience contributes to the growing evidence base supporting observational approaches in appropriately selected cases. Larger prospective studies incorporating molecular and genetic characterization remain essential to identify reliable predictors of lesion behavior and to optimize patient selection for conservative versus surgical management.

Introduction

Contemporary obstetric practice has fundamentally altered the detection and management of pediatric adrenal pathology. Routine prenatal ultrasonography now identifies masses that historically remained undetected until symptomatic presentation later in childhood. Furthermore, improved antenatal detection inevitably identifies lesions that would never have become clinically apparent, a form of overdiagnosis that underscores the importance of selective, risk-adapted management to avoid exposing infants to unnecessary diagnostic and therapeutic interventions. This diagnostic advancement has created therapeutic uncertainties, as traditional oncological principles emphasizing immediate pathological diagnosis increasingly conflict with emerging evidence supporting observational management in selected neonatal cases. Perinatal adrenal masses encompass a broad spectrum ranging from benign conditions to malignant neoplasms. Adrenal hemorrhage, the most common benign etiology, occurs in approximately 0.02–0.05% of live births and typically demonstrates progressive involution on serial imaging. Neuroblastoma represents the most frequent malignant diagnosis, accounting for a significant proportion of adrenal masses that persist or enlarge during surveillance [1–5].

Neuroblastoma presents distinct biological behavior in neonatal populations compared to older children. Despite representing the most common extracranial solid malignancy in childhood, neonatal neuroblastoma frequently exhibits spontaneous regression through incompletely understood mechanisms involving apoptosis or cellular differentiation [6–9]. This unique characteristic has prompted reassessment of traditional management paradigms that emphasized early surgical resection [10–12].

Distinguishing neuroblastoma from benign conditions poses diagnostic challenges in the perinatal period [11, 13]. Ultrasound characteristics provide initial guidance: neuroblastoma typically appears as solid or mixed echogenicity masses with increased vascularity on Doppler examination and demonstrates growth over time, whereas adrenal hemorrhage evolves from echogenic to progressively hypoechoic with involution [2, 14–16]. However, imaging features overlap substantially, limiting definitive radiological diagnosis [17–19]. Magnetic resonance imaging offers superior soft tissue characterization without radiation exposure, enabling differentiation between hemorrhagic and solid components [20]. 123-I-Metaiodobenzylguanidine (MIBG) scintigraphy demonstrates high specificity for neuroblastoma in older children but has reduced sensitivity in neonates, with less than 70% of perinatal neuroblastomas demonstrating MIBG avidity [21]. Biochemical markers including urinary catecholamine metabolites show elevated levels in over 90% of neuroblastoma cases in older children, but their reliability in the neonatal period remains uncertain [22, 23]. The SIOPEN (International Society of Pediatric Oncology Europe – Neuroblastoma) consortium conducted the largest observational study addressing this clinical dilemma, enrolling 128 newborns and infants ≤ 90 days with adrenal masses ≤ 5 cm detected by prenatal or neonatal ultrasound. Patients underwent surveillance with serial ultrasound. Surgery was reserved for lesions demonstrating > 40% volume increase or persistence beyond 48 weeks. Approximately one-third ultimately required surgical intervention. Critically, no deaths resulted from delayed diagnosis or observation. This landmark study provided robust evidence supporting selective conservative management [8].

Building on this evidence base, we implemented a conservative surveillance protocol at our tertiary referral center. This report describes our institutional experience over a 10-year period. Our cohort aims to contribute to the collective experience supporting conservative approach and to describe the practical implementation of surveillance protocols in routine clinical practice.

Patients and methods

Study design and patients’ selection

We conducted a single-institution retrospective case series analysis of all patients diagnosed with perinatal adrenal masses at Fondazione Policlinico Universitario Agostino Gemelli IRCCS, Rome, Italy, between June 2015 and May 2025. Our center functions as a regional referral hub for complex neonatal conditions, receiving referrals from surrounding hospitals when perinatal adrenal masses are detected.

Eligible patients were those with adrenal mass ≤ 5 cm in maximal diameter, detected during prenatal screening or within six months of birth, with complete clinical and imaging documentation and a minimum follow-up of six months. Patients with masses detected after six months of age, incomplete documentation, or follow-up shorter than six months were not included.

Clinical variables collected included patient demographics, gestational age at detection, birth characteristics, imaging evolution, biochemical markers, treatment modalities, and outcomes. We summarized patient characteristics using descriptive statistics.

Follow-up duration ranged from 6 months (minimum inclusion criterion) to 10 years (median: 42 months; interquartile range: 18–84 months). One patient (Patient B, Table 1) was lost to follow-up after 8 months following documented lesion regression. For this patient, all available data through the last documented visit are reported. No data imputation methods were applied. Where specific investigations were not performed (e.g., catecholamine measurement unavailable in one patient due to collection difficulties), this is explicitly stated in the results.

Table 1.

Patient demographic, clinical, and diagnostic characteristics at initial presentation

Patient	Sex	Other pathologies	Time of Diagnosis	Gestational age at detection (weeks)	First-line imaging	NSE at diagnosis (ng/mL)	Side of adrenal mass	Mass maximum diameter (mm)
A	F	LFS	Prenatal	30	Ultrasound	38.1	Left	41
B	M	None	Prenatal	13	Ultrasound	36	Right	26
C	M	None	Postnatal	–	Ultrasound	19.9	Right	18
D	F	None	Postnatal	–	Ultrasound	28.5	Right	10
E	F	None	Prenatal	39	Ultrasound	84.3	Right	43
F	M	None	Prenatal	34	Ultrasound	55.2	Right	25
G	M	Dermal sinus	Postnatal	–	MRI	22.1	Right	18
H	M	None	Postnatal	–	Ultrasound	43.1	Left	20
I	F	None	Prenatal	27	Ultrasound	59.4	Left	14

NSE = neuron-specific enolase; M = male; F = female; LFS: Li-Fraumeni syndrome; MRI: Magnetic Resonance Imaging

Ethical considerations

The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of Università Cattolica del Sacro Cuore (protocol DIPUSVSP-09-12-2582). Informed consent was obtained from parents or legal guardians of all patients for clinical management and for retrospective analysis of anonymized clinical data. Written informed consent for publication was obtained from families. All data were anonymized, and access was restricted to the research team to maintain confidentiality.

Imaging protocols and surveillance strategy

Abdominal ultrasound served as the primary surveillance modality for all patients. Examinations were performed by experienced pediatric radiologists using high-frequency linear array transducers (7–12 MHz for superficial structures; 3–5 MHz for deeper visualization). Each examination included: tri-planar measurements (maximum length, width, and anteroposterior diameter); volume calculation (using the ellipsoid formula V = 0.523 × L × W × H); echogenicity characterization (categorized as solid, cystic, or mixed); color Doppler assessment (to assess internal vascularity and flow patterns); evaluation of adjacent structures (assessment for organ displacement or invasion).

Surveillance intervals were as follows: month 1: every 2 weeks; months 2–6: every 4 weeks; following demonstrated regression: every 3 months; for stable or enlarging lesions: monthly evaluation continued through 12 months. Progression was defined as a volume increase exceeding 40% compared to baseline measurement on two consecutive examinations separated by at least 14 days, or development of new solid components within previously cystic lesions. Indications for advanced imaging, such as magnetic resonance imaging or computed tomography (MRI/CT) were: equivocal ultrasound findings regarding solid versus cystic components; suspected solid mass requiring superior soft-tissue resolution; rapid growth pattern suggesting aggressive biology; pre-surgical planning and staging evaluation. MRI examinations utilized T1-weighted, T2-weighted, and diffusion-weighted sequences. Contrast administration and sedation were avoided when possible to minimize risks in neonates. CT was reserved for urgent evaluations or when MRI was contraindicated or unavailable.

123-I-MIBG scintigraphy was performed only when neuroblastoma remained the leading differential diagnosis after first- and second-line imaging, primarily to assess disease extent (including bone marrow involvement) prior to potential surgical intervention.

Biochemical assessment

All patients underwent neuron-specific enolase (NSE) measurement at initial evaluation. NSE was selected as it is widely used as a tumor marker for neuroblastoma and other neural crest-derived tumors, though its specificity in neonates is limited due to potential elevation in various non-malignant conditions. NSE was measured in serum using electrochemiluminescence immunoassay (ECLIA). The threshold of > 25 ng/mL used in clinical decision-making reflects age-adjusted reference values for neonates. This threshold was chosen to reflect age-adjusted reference values for neonates, in whom physiological NSE levels substantially exceed adult reference ranges due to erythrocyte lysis and neonatal neuronal tissue abundance [24]; this is distinct from the institutional upper reference limit of 16 ng/mL applicable to older children and adults. Urinary catecholamine metabolites (vanillylmandelic acid, homovanillic acid) or plasma metanephrines were measured when technically feasible, acknowledging significant collection difficulties in young infants, particularly in outpatient settings.

Treatment decisions and surgical indications

Surgical resection was recommended in the following cases: documented progression > 40% volume increase on serial imaging meeting progression criteria; volume stability after 12 months of observation; clinical symptoms; high-risk radiological features/strong radiological suspicion of malignancy with concerning characteristics.

All surgical decisions were made through multidisciplinary discussion involving pediatric oncology, pediatric surgery, and diagnostic radiology. Families received comprehensive counselling regarding rationale for observation versus immediate intervention, natural history of disease and spontaneous regression phenomenon in perinatal neuroblastoma.

Study outcomes and data analysis

As a retrospective descriptive series, this study did not pre-specify formal primary or secondary endpoints. The descriptive outcomes documented were: (a) natural history of lesions: proportion demonstrating regression, stability, or progression during surveillance; (b) surgical intervention rate and indications; (c) final diagnoses (pathological or clinical); (d) perioperative complications in surgically treated patients; (e) recurrence or disease progression in conservatively managed patients; (f) overall survival.

Given the small sample size (n = 9), only descriptive statistics were employed. Categorical variables are presented as absolute numbers with percentages. Continuous variables are reported as medians with ranges or interquartile ranges where appropriate. No formal hypothesis testing, confidence interval estimation, or comparative statistical analyses were performed due to insufficient sample size to achieve meaningful statistical power. Even descriptive comparisons between subgroups (e.g., neuroblastoma versus benign lesions) would be unreliable given the limited numbers.

Results

Patient demographics and baseline characteristics

Our series comprised five males and four females. The diagnosis of adrenal lesion was prenatal in five cases and postnatal in four. Eight patients received initial diagnosis using color Doppler ultrasound to define morphology, dimensions, echogenicity and vascular flow assessment. MRI was the first imaging modality in one case, where the adrenal lesion was incidentally discovered during diagnostic investigations for dermal sinus (Table 1).

Surveillance strategy and advanced imaging

Our institutional approach emphasized serial imaging surveillance over immediate surgical intervention. Surveillance intervals during the first six months ranged from two to four weeks depending on lesion characteristics and clinical stability, then extending to longer intervals (every 3 months) for patients with regression of the lesion. For cases demonstrating stability or progression, ultrasound scans continued monthly through a year, at which point a multidisciplinary decision was made regarding surgical resection, depending on the evolution in dimensions of the adrenal mass.

CT and MRI were used as second-line imaging for complex lesions requiring superior soft tissue resolution or to confirm initial radiological findings. CT was used in two cases, MRI in six cases (Table 2). These modalities were employed as single examinations, except for one patient who underwent both prenatal and postnatal MRI to evaluate lesion evolution. In one case, ultrasound was the only imaging modality utilized throughout the entire follow-up. 123-I-MIBG scintigraphy was performed only when neuroblastoma remained suspected after first- and second-line imaging. Six patients underwent scintigraphy, resulting in positive uptake in five cases (four neuroblastomas and one adrenal carcinoma). In one patient, the examination not only confirmed adrenal neuroblastoma but also revealed bone marrow metastases, subsequently confirmed by diagnostic biopsy and histological evaluation.

Table 2.

Imaging modalities, radiological diagnoses, lesion evolution, and management outcomes

Patient	Second line imaging	MIBG scan	Radiological suspicion	Evolution on ultrasound follow-up	Treatment	Outcome
A	CT	Yes	Adrenal carcinoma	Partial regression	Lesion excision	Adrenocortical carcinoma, confirmed; disease-free at last follow-up
B	MRI	Yes	Neuroblastoma	Partial regression	Lost at follow-up	Partial regression documented at last visit
C	MRI	Yes	Neuroblastoma	Increase	Surrenectomy	Neuroblastoma; disease-free
D	MRI	Yes	Neuroblastoma	Stability	Surrenectomy	Neuroblastoma confirmed; disease-free
E	CT	Yes	Neuroblastoma	Stability	Surrenectomy	Neuroblastoma confirmed; disease-free
F	MRI	No	Adrenal hemorrhage	Complete regression	None	Clinical diagnosis; no recurrence during available follow-up
G	None	No	Adrenal cyst	Partial regression	None	Clinical diagnosis; no recurrence during available follow-up
H	MRI	Yes	Adrenal cyst	Partial regression	None	Clinical diagnosis; stable at last follow-up
I	MRI	No	Adrenal hemorrhage	Partial regression	None	Clinical diagnosis; no recurrence during available follow-up

Partial regression: >20% volume reduction without complete sonographic resolution; complete regression; complete sonographic disappearance of the lesion

CT = Computed Tomography; MIBG = Metaiodobenzylguanidine scintigraphy; MRI = Magnetic Resonance Imaging

Biochemical markers

NSE was measured in all nine patients; seven showed elevations above the institutional reference range (> 25 ng/mL). However, NSE elevation occurred in both neuroblastoma cases (3 of 4) and benign lesions (4 of 5), precluding diagnostic discrimination in individual patients.

Assessment of blood catecholamine or urinary metanephrine was performed in eight of nine cases. Elevations were noted in two patients, both of whom were subsequently confirmed as neuroblastoma by histopathology. The remaining six patients with catecholamine measurements showed normal values.

Clinical outcomes and surgical interventions

Given the small sample size of this case series (n = 9), the following proportions should be interpreted with caution and are not intended to reflect population-level outcomes. After one year of radiological follow-up, lesion regression occurred in six cases (66.7%); stability was observed in two cases (22.2%); and progression in one (11.1%). Surgery was performed in four patients: the two cases demonstrating stability after one year of surveillance, one case showing progression, and one case of suspected adrenal carcinoma despite volumetric reduction. The two patients with ultrasound stable lesions underwent surgical resection after 12 months of surveillance; in both cases, the indication for surgery was persistence beyond the protocol threshold, combined with persistently solid echogenicity and internal vascularity on Doppler assessment raising concern for neuroblastoma; volumetric stability in the context of these radiological features was not considered sufficient justification for continued observation. The subsequent histological results confirmed the initial radiological suspects in all four cases (three neuroblastomas, one adrenal carcinoma). No significant surgical complications were encountered in any case. Regarding Patient A suspicion of adrenocortical malignancy arose not from imaging characteristics alone but from the combination of interval growth on serial imaging, and the identification of a maternal TP53 germline mutation consistent with Li-Fraumeni syndrome; definitive diagnosis of adrenocortical carcinoma was confirmed on histopathological examination following surgical resection.

Among the five patients managed non-surgically (all demonstrating lesion regression), none subsequently presented with clinical findings suggesting malignancy or recurrence during follow-up, even if is necessary to underline that three patients have radiological controls of less than 12 months, limiting conclusions about their long-term disease-free status.

Median follow-up was 42 months (range: 6–128 months). Three patients had follow-up under 12 months, four between 12 and 60 months, and two 10 years. One patient (Patient B) was lost to follow-up after achieving partial regression at 8 months; attempts to contact the family were unsuccessful.

No mortality was recorded in this series. All surgically treated patients remain disease-free. No recurrence was observed during available follow-up; however, three patients have follow-up durations of less than 12 months, which is insufficient to draw conclusions about long-term disease-free status in these individuals.

Discussion

This study is best interpreted as a descriptive case series documenting institutional experience over a 10-year period; the findings are hypothesis-generating and should not be extrapolated beyond this carefully selected cohort. The incidence of neuroblastoma in our cohort (44.4%) reflects referral patterns to a tertiary care center and falls within the range reported in the literature, where the percentage of pathologically confirmed neuroblastoma among perinatal adrenal masses varies widely, from approximately 30% to over 80%, depending on the study population and referral patterns [8, 18].

Given the small sample size (n = 9), all proportions cited in this section should be interpreted with caution and are not intended to be generalized beyond this cohort. The predominance of spontaneous regression observed in our series is 66.7%, with one case in which the lesion became no longer detectable on ultrasound within the first year of follow-up. The phenomenon of lesion regression is well known and particularly evident in the perinatal period. Research has identified several potential mechanisms including neurotrophin deprivation affecting nerve growth factor in the tumor microenvironment, immune-mediated cell killing by anti-neuroblastoma antibodies or natural killers (NK), telomere shortening with low or absent telomerase levels, and epigenetic factors. Despite significant advances, the exact molecular mechanisms underlying spontaneous regression remain incompletely understood [25].

The management of Case A warrants specific consideration due to the rare pathological diagnosis of adrenal cortical carcinoma (later confirmed also by an external specialized laboratory). Although initial multidisciplinary approach indicated a standard conservative management—prompted by an early reduction in lesion size—the clinical strategy was re-evaluated following a period of volumetric stabilization. MIBG positivity in this case is atypical, as adrenocortical carcinoma does not ordinarily express the norepinephrine transporter targeted by MIBG; this finding may reflect non-specific or false-positive uptake or could represent an unusual biological feature of this tumor. Furthermore, on ultrasound and CT, the lesion presented with a well-marginated structure, internal neovascularization, and peripheral calcifications. This heterogeneous pattern created diagnostic ambiguity between a typical neonatal neuroblastoma and atypical adrenal tumors. The decision to proceed to surgical resection despite documented volumetric reduction illustrates a critical principle: radiological trend alone does not determine management. In this case, the genetic context (TP53 germline mutation), and the multidisciplinary team assessment collectively drove the decision to operate, overriding the apparent radiological response. After complete staging, given the complete surgical resection and the absence of local or distant metastases, no further local or systemic treatment was performed, in line with the cohort described in literature [13]. This case serves as a reminder that conservative management protocols must incorporate clinical context holistically and should not be managed based on imaging alone.

The presence of MIBG uptake in all four neuroblastoma cases aligns with published data, but our sample size precludes assessment of sensitivity, specificity, or predictive values. The limited utility of biochemical markers in our series warrants acknowledgment. While NSE elevation occurred in most neuroblastoma cases, similar elevations were observed in benign conditions. NSE can be elevated in various neonatal conditions including perinatal hypoxia, intracranial hemorrhage and infections, reducing its diagnostic specificity [26]. The assessment of blood catecholamine or urinary metanephrine levels may have uncertain value in achieving definitive diagnosis [27].

Our institutional experience, combined with the evidence generated by the SIOPEN study, has led us to revise our surveillance protocol. We now reserve MRI for cases with ultrasound indeterminate features and apply NSE sampling selectively rather than routinely, recognizing its limited diagnostic performance in the neonatal period. The decision for surgical intervention evolved during the study period, with earlier cases more likely to undergo resection based on traditional oncological principles emphasizing tissue diagnosis. Our outcomes were favorable, with no perioperative complications observed in the four patients who underwent resection, though the small sample limitations. To optimize conservative management, our clinical experience suggests that clinicians might prioritize initial counseling and the establishment of long-term support structures, to ensure family engagement. It is essential to provide a detailed explanation of the clinical context—specifically, that the mass is statistically likely to regress—while maintaining the diagnostic classification of a potentially malignant growth. Furthermore, providing families with explicit intervention thresholds and transparent surveillance protocols may mitigate the psychological burden of uncertainty of a wait-and-see strategy.

Although psychosocial outcomes were not formally assessed in this study, our clinical experience suggests that parental anxiety represents a significant and frequently underappreciated dimension of conservative management. Parents confronted with the recommendation to observe a potentially malignant mass require thorough initial counseling and ongoing support throughout the surveillance period. Future studies incorporating validated instruments to assess parental anxiety and family quality of life would provide important objective data on this aspect of conservative management.

Study limitations

Several methodological limitations must be acknowledged. Five patients managed conservatively did not undergo histopathological examination; their diagnoses remain clinical and radiological, and the possibility of diagnostic misclassification, however small, cannot be entirely excluded. Urinary catecholamine assessment was not obtained in one patient due to collection difficulties, representing an incomplete biochemical evaluation. MIBG scintigraphy was performed in only six of nine patients; although selection criteria are described, potential selection bias and its effect on diagnostic conclusions should be acknowledged. Germline genetic testing was not performed in most patients; absence of molecular characterization in patients with confirmed neuroblastoma represents a relevant gap. Three patients have follow-up of less than 12 months, which is insufficient to draw conclusions about long-term disease-free status; statements regarding absence of recurrence apply only to available follow-up periods. Finally, the persistence of all confirmed or suspected neuroblastomas beyond one year of age in our cohort, without any spontaneous regression among this subgroup, is likely attributable to the small sample size and possible referral bias toward more complex or persistent lesions, rather than representing a genuine departure from published regression rates.

Additional limitations include retrospective design, heterogeneous follow-up duration, and the referral bias inherent to a tertiary centre, which may have enriched our cohort with more complex or diagnostically challenging cases compared to community settings.

Conclusion

In this retrospective case series of nine infants with perinatal adrenal masses managed at a single tertiary center over 10 years, selective conservative management with serial ultrasound surveillance yielded favorable outcomes, with spontaneous regression observed in six of nine lesions. Temporal evolution of lesion volume proved more informative for clinical decision-making than initial imaging characteristics. The adrenocortical carcinoma case illustrates that radiological trend alone does not always determine management, and that clinical context, including genetic predisposition, must guide decisions holistically. Although our small cohort precludes generalizable conclusions, this institutional experience contributes to the growing evidence base supporting observational approaches in appropriately selected cases. Larger prospective studies incorporating molecular and genetic characterization remain essential to identify reliable predictors of lesion behavior and to optimize patient selection for conservative versus surgical management.

Abbreviations

123-I-MIBG

123-I-Metaiodobenzylguanidine

SIOPEN

International Society of Paediatric Oncology Europe – Neuroblastoma

MRI

Magnetic Resonance Imaging

CT

Computed Tomography

NSE

Neuron-specific Enolase

ECLIA

Electrochemiluminescence Immunoassay

LFS

Li-Fraumeni syndrome

NK

Natural Killer (cell)

Author contributions

S.M. (Stefano Mastrangelo) and A.R. (Antonio Ruggiero) : literature search, writing—original draft, writing—reviewing and editing. F.F., G.A, A.Ro, D.T, L.N. and P.M.: conceptualization, supervision, writing—reviewing and editing. All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by the European Neuroblastoma Association (ENEA) Onlus.

Data availability

The data that support the findings of this study are available from the Corresponding Author, A.R., upon reasonable request.

Declarations

Ethics approval

The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of Università Cattolica del Sacro Cuore (DIPUSVSP-25-06-2166).

Consent to participate

Informed consent was obtained from parents or legal guardians of all patients for clinical management and for retrospective analysis of anonymized clinical data.

Consent for publication

Written informed consent for publication was obtained from families of all patients. All data have been anonymized.

Competing interests

The authors declare no competing interests.