Resolved GH excess and arginine vasopressin deficiency following pituitary apoplexy

Abstract

A 10-year, 8-month-old male presented with polyuria, polydipsia, and secondary nocturnal enuresis for 4 months following a severe headache. Initial fasting studies raised concern for partial arginine vasopressin deficiency (AVP-D). Brain magnetic resonance imaging revealed a possible pituitary adenoma vs artifact. His height was >99th percentile; he was prepubertal and growth charts demonstrated growth velocity >7.5 cm/year for the past 2 years. Additional testing showed markedly elevated IGF-1 and IGF-binding protein 3. He passed a water-deprivation test, ruling out AVP-D. He underwent a GH suppression test that revealed no suppression, suggesting GH excess. Repeat magnetic resonance imaging delineated a cystic pituitary lesion, most consistent with Rathke cleft cyst or hemorrhagic adenoma. Surgery was not recommended because the hematoma obscured visualization of any underlying tumor and evacuation could result in indiscriminate removal of the native gland. Serial imaging demonstrated continued retraction of the hematoma and tumor complex over 10 months. His AVP-D resolved, growth velocity slowed, and growth factors normalized over 13 months without intervention. Tall stature with increased growth velocity in prepubertal children should raise concern for GH excess. GH excess secondary to GH-secreting pituitary adenomas may spontaneously resolve if pituitary apoplexy occurs.

Introduction

GH excess is a rare condition in pediatrics [1]. If left untreated, GH excess can lead to gigantism and metabolic abnormalities including diabetes and hypertension [2]. Treatment of a GH-secreting adenoma typically involves surgical resection and/or administration of somatostatin analogues [2]. We present an interesting case of GH excess (and transient arginine vasopressin deficiency (AVP-D) in a pediatric patient due to a pituitary adenoma that resolved spontaneously following pituitary apoplexy.

Tall stature with increased growth velocity in prepubertal children should raise concern for GH excess, even in the setting of tall familial genetics. Close height monitoring and early referral to endocrinology can assist with early detection of potential intracranial pathology. In children, GH excess most often occurs in the context of an underlying diagnosis of neurofibromatosis type 1 (NF1) or a sporadic pituitary adenoma. Although rare cases of GH excess in the setting of NF1 can remain stable and may not require intervention, GH-secreting tumors typically necessitate treatment with surgical resection. If resection is not feasible, treatment with somatostatin analogues or GH-receptor antagonists may be considered to prevent complications, including gigantism, diabetes, hypertension, left ventricular hypertrophy, arrhythmias, and obstructive sleep apnea [2]. These treatments can be associated with side effects, including cardiac arrhythmia, edema, abdominal pain, biliary tract disease, anemia, and arthralgia [3]. This case is notable for the spontaneous resolution of GH excess after hemorrhagic infarction of the suspected underlying adenoma, raising the possibility that a watchful waiting approach could be considered in carefully selected cases.

Case presentation

A 10-year, 8-month-old, previously healthy male presented to endocrinology for polyuria, polydipsia, and secondary nocturnal enuresis. He was fully continent by age 4 years. Four months before presentation, he had experienced a sudden onset of a severe headache that lasted for 3 days and was treated as migraine. The headache resolved, without recurrence. A few weeks following this event, he developed bedwetting, which was initially attributed to drinking more soda, but this persisted despite limiting fluid intake in the evening and cutting back on all soda and juice. When measured, he was drinking and urinating 4 to 6 L every day (urine output 3.1-4.7 mL/kg/hour; or 2531-3797 mL/m²/day). He reported good energy levels and denied weight loss, polyphagia, weakness, recurrent abdominal pain, and constipation. He was not taking any medications. He had recently noted the onset of puberty including body odor and pubic hair.

On examination, he was noted to have tall stature (>99th percentile for height) (Fig. 1), out of proportion to familial genetics (mid-parental height at 90th percentile), and weight within normal range. He had Tanner I testes. Review of his growth chart demonstrated increased growth velocity starting at age 6 years.

Figure 1.

Growth chart. Star represents MRIs, initial was at outside hospital, subsequent were at our facility growth velocity >10 cm/y from age 6 to 10 years (historic data). Growth velocity 8 cm/y from age 10 to 10 years, 8 months (latter height measured at our endocrinology clinic). Growth velocity 4.5 cm/y from age 10 years, 8 months to 11 years, 5 months (subsequent follow up with outside endocrinologist). IGF-1: 515 ng/mL at age 11 years, 3 months. Growth velocity 2.7 cm/y from age 11 years, 5 months to age 11 years, 8 months; IGF-1: 343 ng/mL at age 11 years, 8 months.

Diagnostic assessment

Evaluation completed before initial visit included fasting laboratory tests (after 12 hours), which showed: sodium: 141 mmol/L (reference range 136-145 mmol/L), serum osmolality (osm): 294 mOsm/kg (reference range 275-300 mOsm/kg), urine osm: 463 mOsm/kg (reference range 50-1200 mOsm/kg), and copeptin <1.8 pmol/L (reference range <14.5 pmol/L), which raised concern for partial AVP-D. A brain magnetic resonance imaging (MRI) scan was completed, which revealed left pituitary lobe larger than right, concerning for possible microadenoma vs artifact. The size of the lesion could not be confirmed by our radiologist as dental braces were creating an artifact that obscured the lesion and there were no thin-cut images of the sella.

Hence, a repeat MRI scan after removal of dental braces was recommended. Subsequent MRI images were obtained at our institution on 3-Tesla MRI scanners. Thin-cut sequences through the sella were obtained before and after administration of gadolinium-contrast administration.

Further workup was completed after endocrine consultation (Table 1), and notable findings included elevated growth factors (IGF-1—483.1 ng/mL [63.1 nmol/L; reference range 81.3-255.3 ng/mL or 10.6-33.3 nmol/L for Tanner I]) (+5.41 SD for Tanner I), IGF-binding protein 3 (IGF-BP3)—5503 ng/mL (719.4 nmol/L) (reference range 2709-5260 ng/mL or 354.1-687.6 nmol/L for Tanner I) (>2 SD For Tanner I), normal thyroid function, and normal prolactin.

Table 1.

Testing after initial endocrine consultation

	Reference range	Value	Z-score
IGF-1	81.3-255.3 ng/mL (10.6-33.3 nmol/L) (for tanner I)	483.1 ng/mL (63.1 nmol/L)	+5.41 for Tanner I
IGF-BP3	2709-5260 ng/mL (354.1-687.6 nmol/L)	5503 ng/mL (719.4 nmol/L)	>+2 SD for Tanner I
TSH	0.510-4.0 μIU/mL (0.51-4.0 IU/L)	1.520 μIU/mL (1.52 IU/L)
Free T4	1.1-2.0 ng/dL (14.1-25.7 pmol/L)	1.6 ng/dL (20.5 pmol/L)
Prolactin	1.9-14.5 ng/mL (1.9-14.5 ng/mL)	16.7 ng/mL (16.7 ng/mL)

Abbreviation: IGF-BP3, IGF binding-protein 3.

Four weeks later, he was admitted for a formal water deprivation test and a repeat brain MRI scan. This evaluation was notable for normal urine concentration after 8 hours of fasting (urine concentrated to >600 mOsm/L twice and sodium remained 141-143 mmol/L). Repeat growth factors remained elevated (IGF-1 548.3 ng/mL or 71.6 nmol/L [+5.87 SD; reference range 81.3-255.3 ng/mL or 10.6-33.3 nmol/L for Tanner I], IGF-BP3-5541 ng/mL [724.4 nmol/L]; >2 SD; reference range 2709-5260 ng/mL or 354.1-687.6 nmol/L for Tanner I). Cortisol was nonrobust (6.1 μg/dL or 168.2 nmol/L; reference range for 8 Am value >10 μg/dL or >275.8 nmol/L), so an ACTH stimulation test was completed and resulted as normal (cortisol at 0 minutes 5.6 μg/dL or 154.4 nmol/L, cortisol at 20 minutes following 1 mcg cosyntropin 23.3 μg/dL or 642.7 nmol/L; goal >18 μg/dL or >496.5 nmol/L at 20 minutes following cosyntropin). He underwent GH suppression test with 75 g of Glucola, which showed elevated GH levels (Table 2). Repeat MRI, performed following removal of his orthodontia, demonstrated a 1.0 × 0.7 cm tumor (which had previously measured 1.4 × 0.8 cm) intermixed with susceptibility artifact suggesting old hemorrhage, as well as inability to definitively visualize the posterior pituitary bright spot. Overall, it was suspected his imaging findings were most consistent with hemorrhagic adenoma, with Rathke cleft cyst felt to be far less likely given the presence of associated hemorrhagic signal changes and documented decrease in size. Genetic testing for overgrowth syndromes was negative.

Table 2.

OGTT to assess GH suppression

Treatment

The neurosurgery team considered surgical resection but it was not recommended as there was concern for potential adverse sequalae following resection due to hematoma obscuring the full extent of the lesion. If attempted, surgical resection may have resulted in incomplete removal of the tumor, or inadvertent removal of the normal gland, resulting in permanent hypopituitarism. Hence, the patient was offered somatostatin analogues. Baseline echocardiogram was normal. By this time, the patient's AVP-D had resolved and his growth velocity was starting to spontaneously slow and return to a normal rate. As such, his family opted for close monitoring rather than starting octreotide.

Outcome and follow up

The patient was monitored clinically and through serial laboratory tests and MRI. Over the following year, follow-up MRI studies revealed gradual resolution of hematoma and tumor complex, with the overall lesion size measuring 0.6 × 0.3 cm by 10 months following initial presentation (Fig. 2). In addition, his growth velocity and growth factors declined and eventually normalized. At present, he is now in early puberty with appropriate IGF-1 levels. He will continue requiring annual MRI and pituitary screening laboratory tests, including IGF-1, IGF-BP3, thyroid function, cortisol, and testosterone; and monitoring linear growth every 6 months until the end of puberty. Long-term monitoring for cardiovascular and metabolic complications associated with GH excess is also warranted. However, his risk of recurrence is low given negative genetic testing.

Figure 2.

Serial MRI. Initial at Outside Hospital: 1.4 × 0.8 cm: image not available. 1 month later: 1.0 × 0.7 cm (A). 4 months later: 0.7 × 0.4 cm (B). 10 months later: 0.6 × 0.3 cm (C).

Discussion

This is a unique presentation of previously unidentified GH excess, associated with a GH-secreting adenoma, presenting with clinical, transient suspected AVP-D (although not fully confirmed by biochemical investigation) and sudden, severe headache resulting from pituitary apoplexy, distinct from GH excess. Following hemorrhagic infarction of the adenoma, the patient's growth factor levels normalized spontaneously. We suspect the patient's slow-growing GH-secreting adenoma spontaneously hemorrhaged, causing severe headache and transient AVP-D due to intracranial pressure change, and as the lesion involuted the GH excess resolved. Spontaneous resolution of GH excess, without therapeutic intervention, has not been documented previously in the pediatric literature, although there have been a few reports of NF1-related GH excess remaining stable over time [4, 5]. There are few cases (<35 case reports to our knowledge) reported in the adult literature of pituitary apoplexy and spontaneous resolution of GH excess [6-8]. Many of these patients had an inciting cause of the apoplexy and for several, the initial presenting symptom was headache without known preexisting pituitary adenoma. This is similar to our patient as GH excess was not detected for several years before presentation. In the adult reports, while acromegaly resolved with pituitary apoplexy, most patients eventually had one or more pituitary deficiencies. To our knowledge, this presentation has not been reported in a pediatric patient thus far. Close linear growth monitoring and early evaluation of tall stature, even in the setting of tall familial genetics can lead to earlier diagnosis of GH excess and reduced risk of complications of untreated GH excess as further discussed later.

In rare cases, observation may be a reasonable strategy, especially if surgical resection is not advisable. However, untreated GH excess can lead to several complications including gigantism, diabetes, hypertension, left ventricular hypertrophy, arrhythmias, obstructive sleep apnea, colonic polyps, and cancer [2, 9]. It may also be associated with increased risk of mortality because of cardiovascular complications [9]. As such, in an unusual circumstance where therapeutic intervention is not initiated for a GH-secreting adenoma, close clinical follow up and monitoring is essential.

Our patient has been taller than the 99th percentile since age 6 years, a finding that had been documented but was presumed to be due to familial genetics. Had there been earlier endocrinological evaluation, the pituitary adenoma potentially could have been detected before his presentation with apoplexy and managed appropriately at an earlier age.

This case demonstrates the importance of early evaluation of tall stature in pediatrics, which can lead to timely diagnosis and management before onset of complications like pituitary apoplexy.

Learning points

• Close height monitoring and early referral to endocrinology leads to earlier detection of intracranial pathology.

• Tall stature, especially with increased growth velocity, in prepubertal children should raise concern for a GH-secreting adenoma, even in the setting of tall familial genetics.

• GH excess may spontaneously resolve, at least temporarily, if pituitary apoplexy occurs.

Contributors

B.M. and S.S. were involved in the diagnosis and management of the patient. B.M. drafted the initial manuscript. C.S. was responsible for the patient's MRI review and surgical opinion. All authors reviewed and approved the final draft.

Current affiliation

Current affiliation for Bhavapriya Mani: Division of Pediatric Endocrinology, University of Minnesota Masonic Children's Hospital, Minneapolis, MN 55454, USA.

Funding

No public or commercial funding.

Disclosures

The authors have no conflicts of interest to disclose.

Informed patient consent for publication

Signed informed consent could not be obtained from the patient or a proxy but has been approved by the treating institution.

Data availability

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.