Journal of Pediatric Ophthalmology and Strabismus

Short Subjects 

Nystagmus in the Diagnosis of Russell Diencephalic Syndrome

Megan Tuohy, MD; Patricia L. Robertson, MD; Francisco Rivas-Rodriguez, MD; Jonathan D. Trobe, MD

Abstract

Russell diencephalic syndrome is a condition in which infants become emaciated in the setting of a decreased or normal caloric intake as the result of a hypothalamic astrocytoma. The diagnosis may be delayed if providers initially attribute the symptoms to a behavioral disorder. The detection of nystagmus, which is present in many patients, may be a critical diagnostic clue. The authors describe two patients in whom the discovery of nystagmus months after the onset of emaciation led to the diagnosis of Russell diencephalic syndrome. [J Pediatr Ophthalmol Strabismus. 2019;56:e79–e83.]

Abstract

Russell diencephalic syndrome is a condition in which infants become emaciated in the setting of a decreased or normal caloric intake as the result of a hypothalamic astrocytoma. The diagnosis may be delayed if providers initially attribute the symptoms to a behavioral disorder. The detection of nystagmus, which is present in many patients, may be a critical diagnostic clue. The authors describe two patients in whom the discovery of nystagmus months after the onset of emaciation led to the diagnosis of Russell diencephalic syndrome. [J Pediatr Ophthalmol Strabismus. 2019;56:e79–e83.]

Introduction

Russell diencephalic syndrome is a rare condition of failure to thrive with emaciation in young children who have a normal or decreased caloric intake and a hypothalamic tumor. First described by Russell1 in 1951 in a series of five infants with unexplained failure to thrive, subsequent reports established that approximately all children become symptomatic in the first 2 years of life and may fail to gain weight appropriately even with adequate caloric intake. Some children have preserved linear growth and hyperkinetic behavior. The tumors are uniformly World Health Organization grade I pilocytic astrocytomas that involve both the optic chiasm and the hypothalamus.1–3 Diagnosis is often delayed.2,4–6 We report two cases in which the discovery of nystagmus was the tip-off.

Case Reports

Case 1

A 13-month-old girl was admitted to the pediatric gastroenterology service for failure to thrive. Her father had been diagnosed as having neurofibromatosis type 1 (NF1) in childhood. Her birth history was normal, with no health concerns before 2 months of age. At 3 months of age, she refused feedings until a formula that she would drink was found. Her birth weight was in the 38th percentile, but it dropped to the 4th percentile by 3 months of age. Between 3 and 8 months of age, her caloric intake was reportedly appropriate, but her weight percentile declined to less than the 1st percentile (Figure 1). At 6 months of age, five café au lait spots and a small, dermal neurofibroma were noted on her skin, which did not fulfill the diagnostic criteria for NF1. At 8 months of age, she began to have intermittent vomiting, variable caloric intake, and a further decline in her weight percentile. At 11 months of age, the patient was evaluated by the pediatric gastroenterology service. She was noted to have a caloric intake below the recommended level, partly attributed to parental error in preparation of her formula. By 1 year of age, her weight percentile continued to fall, which seemed related to behavioral issues with feeding because she sometimes refused to eat and threw food on the floor but ate things she liked. This hypothesis was reinforced by inconclusive results of metabolic/endocrine laboratory studies and an upper gastrointestinal endoscopy.

Growth chart of patient 1 from birth to publication. (A) The figure on the left demonstrates a dramatic decline in weight percentile from birth to 12 of age months with a continual gradual increase in weight percentile between 18 and 24 months after the start of second-line chemotherapy (vinblastine). (B) Her length showed a similar trajectory.

Figure 1.

Growth chart of patient 1 from birth to publication. (A) The figure on the left demonstrates a dramatic decline in weight percentile from birth to 12 of age months with a continual gradual increase in weight percentile between 18 and 24 months after the start of second-line chemotherapy (vinblastine). (B) Her length showed a similar trajectory.

At 12 months of age, the patient's grandmother noticed an intermittent “shaking” of the patient's eyes, prompting an ophthalmologic consultation. The patient was an irritable, thin child who did not engage with examiners. She had a prominent, multivector pendular nystagmus (Video 1, available in the online version of this article). She fixed and followed light and was aversive to it. The patient had a left afferent pupil defect and mild pallor of the left optic disc, indicative of a relative left optic neuropathy. She was unable to sit independently and displayed language delay.

The combination of optic neuropathy and nystagmus suggested a diencephalic lesion. Magnetic resonance imaging of her brain (Figure 2A) disclosed a thickened left retrobulbar optic nerve and a large chiasmal-hypothalamic mass with compression of the temporal lobe and midbrain. The tumor was hypointense on T1-weighted imaging and enhanced heterogeneously after intravenous contrast administration, which are features typical of a pilocytic astrocytoma, the most common brain tumor in patients with NF1. Patchy areas of abnormal fluid-attenuated inversion recovery hyperintensity were present in the deep white matter of the cerebellar hemispheres and corpus callosum, consistent with intramyelinic edema as seen in children with NF1. Given the family history of NF1 and the distinctive imaging abnormalities, a histologic verification by biopsy was deemed unnecessary.

Case 1: Magnetic resonance imaging of the brain. (A) Age 13 months. Sagittal post-contrast T1-weighted image at time of diagnosis and initiation of chemotherapy. It shows a large enhancing mass that is centered in the hypothalamus, involves the optic chiasm, compresses the third ventricle, and pushes the midbrain backward. Signal characteristics are so characteristic of pilocytic astrocytomas that a biopsy was considered unnecessary in the setting of neurofibromatosis type 1. (B) Age 18 months. Sagittal post-contrast T1-weighted image shows tumor growth causing increasing ventriculomegaly that required ventriculo-peritoneal shunting and septostomy (not shown). (C) Age 22 months. Sagittal post-contrast T1-weighted image shows tumor shrinkage. The patient had gained weight with gastrojejunal feedings and was more cheerful. The nystagmus disappeared.

Figure 2.

Case 1: Magnetic resonance imaging of the brain. (A) Age 13 months. Sagittal post-contrast T1-weighted image at time of diagnosis and initiation of chemotherapy. It shows a large enhancing mass that is centered in the hypothalamus, involves the optic chiasm, compresses the third ventricle, and pushes the midbrain backward. Signal characteristics are so characteristic of pilocytic astrocytomas that a biopsy was considered unnecessary in the setting of neurofibromatosis type 1. (B) Age 18 months. Sagittal post-contrast T1-weighted image shows tumor growth causing increasing ventriculomegaly that required ventriculo-peritoneal shunting and septostomy (not shown). (C) Age 22 months. Sagittal post-contrast T1-weighted image shows tumor shrinkage. The patient had gained weight with gastrojejunal feedings and was more cheerful. The nystagmus disappeared.

A gastrostromy tube was placed to optimize nutrition. She began a chemotherapy regimen of weekly infusions of carboplatin (175 mg/m2) and vincristine (0.05 mg/kg). Four months later, magnetic resonance imaging of her brain showed tumor growth (Figure 2B) and an obstructive hydrocephalus that required a ventriculo-peritoneal shunt and fenestration of the septum pellucidum. The treatment regimen was changed to weekly vinblastine (0.2 mg/kg/dose).

With nasojejunal tube feedings, the patient's weight reached the 32nd percentile by 18 months of age. Additionally, she made neurodevelopmental progress, the tumor shrunk substantially (Figure 2C), and the nystagmus disappeared (Video 2, available in the online version of this article). The optic neuropathy was unchanged.

Case 2

A healthy boy born at full term of an uneventful pregnancy with a birth weight of 3.7 kg (73rd percentile) gained weight appropriately in the first month of life and met normal developmental milestones. Then, breast feeding slowed and his weight decreased to the 15th percentile. At 3 months of age, he refused breast feeding and sustained a drastic weight loss (< 1st percentile). Nasogastric feeding led to vomiting. In the following 6 months, he failed to gain weight and remained at less than the 1st percentile. Given the pattern of oral aversion and apparent absence of gastroesophageal reflux, gastroenterology consultants concluded that his weight loss was most likely due to behavioral factors. A psychiatrist suggested that his aversion to food intake was at least partly attributable to overzealous feeding.

At 6 months of age, the patient's mother noted ocular oscillations, which prompted magnetic resonance imaging of his brain that showed a large suprasellar mass originating in the hypothalamus (Figure 3A). An ophthalmic examination showed the patient's ability to fixate and follow light, pupils of normal size and reactivity, normal ocular alignment, and profoundly pale optic discs. His eye movements were full, but there was a circular pendular nystagmus of moderate amplitude in both eyes in all fields of gaze (Video 3, available in the online version of this article). There were no signs of NF1 on the skin examination.

Case 2: Magnetic resonance imaging of the brain. (A) Sagittal midline post-contrast T1-weighted sequence at diagnosis shows a large, heterogeneously enhancing mass centered at the hypothalamus/chiasm/suprasellar cistern. It displaces the third ventricle, midbrain, and corpus callosum. (B) Sagittal post-contrast T1-weighted sequence obtained 3 months after diagnosis shows interval growth of the mass. (C) Similar sequence obtained 6 months after diagnosis shows further growth of mass.

Figure 3.

Case 2: Magnetic resonance imaging of the brain. (A) Sagittal midline post-contrast T1-weighted sequence at diagnosis shows a large, heterogeneously enhancing mass centered at the hypothalamus/chiasm/suprasellar cistern. It displaces the third ventricle, midbrain, and corpus callosum. (B) Sagittal post-contrast T1-weighted sequence obtained 3 months after diagnosis shows interval growth of the mass. (C) Similar sequence obtained 6 months after diagnosis shows further growth of mass.

The pathologic diagnosis from an open brain biopsy was BRAF-KIAA 1549 fusion–positive and BRAF-V600E mutation-negative pilocytic astrocytomas, supporting its low-grade nature. The patient was randomized to a regimen of monthly carboplatin in a clinical trial. After two cycles, his head circumference increased beyond the 95th percentile. Magnetic resonance imaging of his brain revealed tumor growth and lateral ventricular expansion consistent with obstructive hydrocephalus (Figure 3B), for which he underwent the placement of a ventriculo-peritoneal shunt and fenestration of the septum pellucidum. He underwent weekly carboplatin and vincristine treatments, but the clinical course was affected by several central line infections, sepsis, development of ascites due to decreased peritoneal absorption of cerebrospinal fluid, and the need for shunt revision. Nine months after starting chemotherapy, he achieved a weight gain to the 8th percentile. However, the tumor volume increased (Figure 3C), so he began a regimen of oral trametinib, a MEK inhibitor, at 0.23 mg daily (0.025 mg/kg).

Discussion

The two cases describe clinical features that are common in Russell diencephalic syndrome. Both infants refused feedings and drastically fell off growth curves for weight, even after receiving an increased caloric intake through nasogastric or nasojejunal tube feedings. When the initial gastrointestinal evaluations revealed no explanation, these phenomena were attributed to inadequate caloric intake related to behavioral issues. It was only when the nystagmus was discovered that magnetic resonance imaging of the brain was performed, disclosing a hypothalamic mass.

A characteristic feature of this syndrome is the failure to gain appropriate weight in the face of adequate caloric intake, as was seen in our patients. This phenomenon has been ascribed to a catabolic state with increased basal energy expenditure requiring more than 200% of usual caloric intake.4 Fleischman et al.2 hypothesized elevated basal hormone levels and incomplete suppression after glucose loading, and possibly dysregulation of somatostatin, ghrelin, and leptin. These alterations are attributed to tumor-induced disruption of relevant metabolic pathways in the anterior hypothalamus of infants, which may be more vulnerable than those of an older child.

Many children with this condition are described as hyperkinetic, hyperalert, and euphoric in the presence of a cachectic appearance, which was true of case 2 but not case 1, who was irritable and sometimes listless.1,5 The absence of hyperkinetic behavior was noted in two patients in one report6 and in 7 of 8 patients in another report.7

Delays in the diagnosis of Russell diencephalic syndrome have been well documented. In two reported series, the average and mean latencies from recognition of failure to thrive to syndrome diagnosis were 12.5 months (11 patients)5 and 11 months (8 patients).6

The discovery of nystagmus was critical to syndrome diagnosis for our two patients. In case 1, the patient's grandmother first noticed it shortly before the ophthalmologic consultation was requested. Neither the patient's parents nor the examining pediatricians had seen it earlier. In case 2, the patient's parents first noted it at 6 months of age, approximately 3 months after his food intake declined and weight loss set in. In a 1972 report, 45 of 48 patients with this syndrome eventually exhibited nystagmus, 25 of whom demonstrated nystagmus (waveform not specified) early in the disease course.7 However, in another series of 8 patients, only 1 had nystagmus.6 In a further report of 3 patients, the only patient who had nystagmus displayed a waveform similar to that of our patients,8 a pendular and circular motion equal in intensity in both eyes and in all gaze positions. This form of nystagmus has been called “multivector.” The eyes oscillate in horizontal and vertical planes with a relatively slow trajectory. Some observers describe the oscillations as resembling those of an “eggbeater” or “windmill.” This nystagmus waveform is typically associated with myelin disorders that affect the diencephalon and brainstem.9 It may be related to seesaw nystagmus, in which the eyes display cycles of vertical and torsional pendular movements that carry one eye up and the other eye down, a waveform characteristic of diencephalic region masses.

Our patients suffered many medical complications that may not have occurred with earlier diagnosis of their tumors. In both cases, chemotherapy did not initially stop tumor growth, and both patients required cerebrospinal diversion for obstructive hydrocephalus. Because the obstruction caused by the tumor was at the foramina of Monro, both lateral ventricles were blocked. A surgical division of the septum pellucidum was required to allow communication between the two lateral ventricles so a single catheter inserted in one ventricle would drain both ventricles. If the tumors had been detected earlier, it is possible that their growth on the initial chemotherapy regimens would have been found before the hydrocephalus developed, allowing the treatment to be changed to a potentially more effective therapy and obviating the shunt procedures.

In case 1, chemotherapy eventually shrank the tumor, and the child sustained weight gain through nasojejunal and gastrojejunal tube feedings that delivered a normal caloric intake. In case 2, the tumor continued to grow on second-line chemotherapy, possibly due to interruptions in therapy from medical complications. However, his weight has modestly increased.

The difficult clinical course for these patients is typical for most patients with this condition. A 1976 study reported that 18 of 20 patients who were not treated died within 1 year of symptom onset.5 Overall survival is better with specific tumor therapy, but the optimal treatment remains uncertain because no controlled trials have been conducted. Extensive surgical resection is limited by the anatomic location of these tumors. Radiation therapy has unfavorable long-term effects on cognition and neuroendocrine function.10 Due to the contraindications to surgery and radiotherapy, children have been treated with various cytotoxic or cytostatic chemotherapies. These regimens, including carboplatin/vincristine (used as first-line therapy in case 1), thioguanine, procarbazine, lomustine, and vincristine, and vinblastine monotherapy (used as second-line therapy in case 1) have achieved tumor response rates from 35% to 52% and 5-year progression-free survival from 39% to 53%.11,12 A report documented that these regimens were successful in 5 of 7 infants who sustained weight gain (80% median gain) and median progression-free survival of 2 years with a 5-year overall survival of 71%.13 A report of 8 Korean patients disclosed similarly beneficial responses.6

Recent progress in understanding the molecular pathogenesis of pilocytic astrocytomas and other pediatric low-grade gliomas has led to the development of targeted agents. The majority of pilocytic astrocytomas not related to NF1 appear to be driven by a single genetic alteration—the KIAA1549-BRAF fusion that activates the MAPKinase (RAS-RAF-MEK-ERK) signaling pathway.14 The hypothalamic/optic pathway in children with NF1 germline mutations also contains activated MAPKinase pathways. The oral MEK inhibitor selumetinib, which targets that pathway, is used in pediatric clinical trials for relapsed low-grade gliomas.15 Trametinib, the MEK inhibitor used in case 2, has demonstrated efficacy against pediatric hypothalamic gliomas in several case series.16,17

References

  1. Russell A. A diencephalic syndrome of emaciation in infancy and childhood. Arch Dis Child. 1951;26.
  2. Fleischman A, Brue C, Poussaint TY, et al. Diencephalic syndrome: a cause of failure to thrive and a model of partial growth hormone resistance. Pediatrics. 2005;115(6):e742–e748. doi:10.1542/peds.2004-2237 [CrossRef]15930202
  3. Poussaint TY, Barnes PD, Nichols K, et al. Diencephalic syndrome: clinical features and imaging findings. AJNR Am J Neuroradiol. 1997;18(8):1499–1505.9296191
  4. Viachopapadopoulou E, Tracey KJ, Capella M, Gilker C., Matthews DE. Increased energy expenditure in a patient with diencephalic syndrome. J Pediatr. 1993:122(6):922–924. doi:10.1016/S0022-3476(09)90021-X [CrossRef]
  5. Burr IM, Slonim AE, Danish RK, Gadoth N, Butler IJ. Diencephalic syndrome revisited. J Pediatr. 1976;88(3):439–444. doi:10.1016/s0022-3476(76)80260-0 [CrossRef] doi:10.1016/S0022-3476(76)80260-0 [CrossRef]1245953
  6. Kim A, Moon JS, Yang HR, Chang JY, Ko JS, Seo JK. Diencephalic syndrome: a frequently neglected cause of failure to thrive in infants. Korean J Pediatr. 2015;58(1):28–32. doi:10.3345/kjp.2015.58.1.28 [CrossRef]25729396
  7. Tosur M, Tomsa A, Paul DL. Diencephalic syndrome: a rare cause of failure to thrive. BMJ Case Rep. 2017;2017.28687692
  8. Addy DP, Hudson FP. Diencephalic syndrome of infantile emaciation. Analysis of literature and report of further 3 cases. Arch Dis Child. 1972;47(253):338–343. doi:10.1136/adc.47.253.338 [CrossRef]5034666
  9. Leigh RJ, Zee DS. Acquired pendular nystagmus and its relationship to visual pathways. In: Gilman S, Herdman WJ, eds. The Neurology of Eye Movements, 5th ed. New York: Oxford University Press; 2015:696–703.
  10. Duffner PK. Long-term effects of radiation therapy on cognitive and endocrine function in children with leukemia and brain tumors. Neurologist. 2004;10(6):293–310. doi:10.1097/01.nrl.0000144287.35993.96 [CrossRef]15518596
  11. Ater Jl, Zhou T, Holmes E, et al. Randomized study of two chemotherapy regimens for treatment of low-grade glioma in young children: a report from the Children's Oncology Group. J Clin Oncol. 2012;30(21):2641–2647. doi:10.1200/JCO.2011.36.6054 [CrossRef]22665535
  12. Lassaletta A, Scheinemann K, Zelcer SM, et al. Phase II weekly vinblastine in chemotherapy-naïve children with progressive low-grade glioma: A Canadian Pediatric Brain Tumor Consortium Study. J Clin Oncol. 2016;34(29):3537–3543. doi:10.1200/JCO.2016.68.1585 [CrossRef]27573663
  13. Gropman AL, Packer RJ, Nicholson HS, et al. Treatment of diencephalic syndrome with chemotherapy: growth, tumor response, and long term control. Cancer. 1998;83(1):166–172. doi:10.1002/(SICI)1097-0142(19980701)83:1<166::AID-CNCR22>3.0.CO;2-U [CrossRef]9655307
  14. Packer RJ, Pfister S, Bouffet E, et al. Pediatric low-grade gliomas: implications of the biologic era. Neuro Oncol. 2017;19(6):750–761. doi:10.1093/neuonc/now209 [CrossRef]
  15. Banerjee A, Jakacki RI, Onar-Thomas A, et al. A phase I trial of the MEK inhibitor selumetinib (AZD6244) in pediatric patients with recurrent or refractory low-grade glioma: a Pediatric Brain Tumor Consortium (PBTC) study. Neuro Oncol. 2017;19(8):1135–1144. doi:10.1093/neuonc/now282 [CrossRef]28339824
  16. Miller C, Guillaume D, Dusenbery K, Clark HB., Moertel C. Report of effective trametinib therapy in 2 children with progressive hypothalamic optic pathway pilocytic astrocytoma: documentation of volumetric response. J Neurosurg Pediatr. 2017;19(3):319–324. doi:10.3171/2016.9.PEDS16328 [CrossRef]
  17. Kondyli M, Ellezam B, Saint-Martin C, et al. Trametinib and dabrafenib for refractory/inoperable pediatric low grade gliomas. Neuro Oncol. 2017;19(Suppl 6):vi185–vi186. doi:10.1093/neuonc/nox168.752 [CrossRef]
Authors

From the Kellogg Eye Center, Ann Arbor, Michigan (MT, JDT); and the Departments of Ophthalmology and Visual Sciences (MT, JDT), Radiology (Neuroradiology) (FR-R), Neurology (PLR, JDT), and Pediatrics (Neuro-oncology) (PLR), University of Michigan, Ann Arbor, Michigan.

The authors have no financial or proprietary interest in the materials presented herein.

Correspondence: Jonathan D. Trobe, MD, University of Michigan, 1000 Wall Street, Ann Arbor, MI 48105. E-mail: jdtrobe@med.umich.edu

Received: May 12, 2019
Accepted: July 25, 2019
Posted Online: December 09, 2019

10.3928/01913913-20190801-01

Sign up to receive

Journal E-contents