Anesthetic considerations for rapid-onset obesity, hypoventilation, hypothalamic dysfunction, and autonomic dysfunction (ROHHAD) syndrome in children

Summary

Rapid-onset obesity, hypoventilation, hypothalamic dysfunction, and autonomic dysfunction is an increasingly common diagnosis in patients who are being seen at tertiary care children’s hospitals. We present two cases of anesthetics from the authors’ own experience in addition to a comprehensive review of the disorder and anesthetic implications.

Introduction

Congenital Central Hypoventilation Syndrome (CCHS) was first reported in the modern era by Mellins et al. in 1970 (1), although ancient treatises mention the entity. Rapid-onset Obesity, Hypoventilation, Hypothalamic dysfunction, and Autonomic Dysfunction (ROHHAD) is a closely related but clinically and genetically distinct entity first described by Ize-Ludlow and colleagues in 2007. They reported 23 children with ROHHAD (2). The later age of onset (median age of 3 years), very dramatic sudden weight gain, and negative POHX2B genotypical sequencing distinguish ROHHAD from CCHS (3). CCHS typically presents in the first few months of life (mean age of 3.5 months) and is associated with positive POHX2B sequencing, but not typically with obesity (4–7). Only about 50 diagnosed cases of ROHHAD have been seen worldwide, while CCHS, although rare (1 : 200 000 live births), is seen in most busy children’s medical centers.

Despite the earlier onset of CCHS, the clinical differences that distinguish CCHS from ROHHAD, and the absence so far of any genetic marker for ROHHAD, these two syndromes are believed to be closely associated functionally and, perhaps, also genetically. Our review of the literature found no articles discussing the anesthetic implications of ROHHAD. We present our experience with two cases of anesthesia for children with ROHHAD.

Case reports

Case 1

A 4.8-year-old, 29-kg, 108-cm girl with a BMI of 24.9 was diagnosed with ROHHAD 5 months prior to surgical admission at Stony Brook University Medical Center for tracheostomy. She had experienced hyperphagia with rapid weight gain beginning 6 months prior to admission. At the age of 4 years, she had developed hypothalamic dysfunction characterized by mild diabetes insipidus (DI) with hypernatremia, and hyperprolactinemia. Her hypoventilation was mild when she was first diagnosed, but overnight noninvasive (mask) ventilation without supplemental oxygen was required within 2 months. Her hypoventilation worsened rapidly, requiring full overnight ventilation with supplemental oxygen to prevent hypercarbia and hypoxemia. She required no ventilatory support while awake. Because of her rapidly progressive hypoventilation and need for long-term ventilatory support while asleep, she was scheduled for tracheostomy.

Her autonomic dysregulation was manifested by dilated pupils, decreased pupillary response to light, cold hands, excessive diaphoresis, and constipation. A comprehensive cardiac evaluation was carried out. Her electrocardiogram (ECG) demonstrated sinus rhythm with a rate of 134 bpm, and an echocardiogram revealed mild right ventricular diastolic dysfunction. Her 24-h Holter monitor revealed reduced beat-to-beat heart rate variability consistent with dysautonomia. She had undergone two prior uneventful general anesthetics, one for left adrenal ganglioneuroblastoma resection and another for laparoscopic left adrenalectomy, approximately 8 months prior to this admission and before being diagnosed with ROHHAD.

The anesthetic plan was to use short-acting drugs that would not depress respiration substantially. A peripheral intravenous (IV) catheter was placed preoperatively. Upon arrival in the operating room, the unpremedicated patient was monitored with pulse oximetry, blood pressure cuff, and five-lead ECG. After preoxygenation, a dexmedetomidine infusion was started at 0.2 mcg·kg⁻¹·h⁻¹. Anesthesia was then induced with ketamine 2 mg·kg⁻¹ and atropine 15 mcg·kg⁻¹ IV. The patient was intubated uneventfully. Anesthesia was maintained with sevoflurane, dexmedetomidine, and additional small IV doses of ketamine. About 10 min after induction, rocuronium, 0.3 mg·kg⁻¹, was given to suppress slight patient movement that occurred despite what was assessed clinically to be an adequate depth of anesthesia. Ventilation was controlled; the ETCO2 ranged from 35 to 55 mmHg.

At the conclusion of the 45-min procedure, neuromuscular blockade was reversed with 50 mcg·kg⁻¹ of neostigmine and 10 mcg·kg⁻¹ of glycopyrrolate IV. The patient was soon ventilating spontaneously and awake. She was transported to the recovery area with supplemental oxygen delivered by mask placed loosely over the tracheostomy tube, with SP02 of 96%. Her ETCO2 was monitored throughout her uneventful 3-h recovery area stay and was lower than 48 mmHg at all times. She spent 24 h in the pediatric intensive care unit and was discharged home on postoperative day 2.

Case 2

A 54-kg, 137-cm (BMI 28.8), 9-year-old girl presented to The Nationwide Children’s Hospital of Columbus, Ohio, for general anesthesia for colonoscopy and placement of colonic manometry catheters. She had undergone several uneventful general anesthetics elsewhere previously and had a complex history with diagnoses including persistent developmental delay, chronic constipation, obesity, generalized weakness, neurogenic bladder, and hypotonia. Her obesity began when she gained 13 kg over a 4-month period during her third year of life, increasing her body mass by 40%.

Despite caloric restriction, her BMI since that time always exceeded 28. She had both rapid and slow pulse rates during clinic visits and was given the diagnosis of ROHHAD at age 6, when she was found also to have DI with chronic hypernatremia. She became dependent on nighttime noninvasive mask ventilation for 1 year prior to admission and had required it also during naps for 4 months prior to admission because of observed low oxyhemoglobin saturations on portable pulse oximetry. The patient’s mother stated that the child had undergone at least four anesthetics prior to her diagnosis with ROHHAD and five since the diagnosis. She stated that the child had received oral midazolam as a premedication for her two most recent anesthetics elsewhere and had no complications or problems. On examination, the child was obese and had diffuse eczema. A gastrostomy tube was present on the anterior abdominal wall. She was slow to respond, but answered simple questions appropriately. She was assessed as having a Mallampati II airway and had normal cardiac and respiratory examinations.

Anesthesia was induced in the unpremedicated child using only sevoflurane in oxygen. The child cooperated in the induction and also in a baseline assessment of endtidal carbon dioxide level (stable at 37 mmHg for a 3-min period of observation immediately prior to induction). During the same baseline period of measurement, the pulse rate varied between 84 and 90, and the respiratory rate was constant between 16 and 20 breaths·min⁻¹. Anesthetic induction was smooth and uneventful. The patient was intubated easily using only sevoflurane. An IV was started, but no medications were given by that route.

The patient’s respiratory and pulse rates, endtidal sevoflurane and carbon dioxide tensions, and transcutaneously measured oxyhemoglobin saturations were measured continuously and recorded every 3 min. Blood pressure was measured and recorded every 3 min during the 45-min anesthetic, with spontaneous ventilation throughout. The endotracheal tube was removed uneventfully in the procedure room at the end of the case. The same variables were measured and recorded in the same manner during the 60-min recovery period; capnography was then accomplished by continuous sampling from a nasal cannula port.

During the anesthetic, blood pressures (systolic, diastolic, and mean) remained within 10% of basal levels, as did respiratory and pulse rates. Even at two MAC sevoflurane (and throughout the anesthetic), endtidal partial pressure of CO2 remained tightly between 34 and 37 mmHg, as it also did throughout the entire recovery period. The other clinical variables measured during recovery showed no unexpected deviations either.

Discussion

ROHHAD is a fairly new diagnosis and identifies a patient population with multiorgan dysfunction. Although the number of children who have this entity is not known to be increasing, the number diagnosed correctly with ROHHAD certainly is. The pediatric anesthesiologist will see increasing numbers of children with ROHHAD. No specific genotypic markers have been identified for ROHHAD; however, CCHS has been associated with frameshift, missense, and nonsense mutations in the PHOX2B sequence (8) following the principles of Mendelian inheritance (9). The mutations appear to code for failure to produce neural pathways involved in the control of respiration. The major implications for the pediatric anesthesiologist are hypoventilation, obesity, and autonomic dysautonomia.

ROHHAD’s etiology is probably multifactorial, given that its course in monozygotic twins is highly variable (10). With its cause currently still unknown, ROHHAD must be distinguished from other syndromes with central hypoventilation such as Prader–Willi syndrome, Beckwith–Wiedemann syndrome, obstructive sleep apnea, and pediatric obesity–hypoventilation syndrome (Pickwickian syndrome). The former two syndromes are confirmed by genetic testing. Obstructive sleep apnea is confirmed by polysomnography, and ROHHAD patients can have sleep apnea symptoms. The other phenotypic features for ROHHAD, however, help with the diagnosis. The Pickwickian syndrome with obesity and alveolar hypoventilation can be more difficult to distinguish from ROHHAD; however, the onset and the presence of other phenotypic features help to disambiguate the two entities. Clinically overlapping cases of CCHS and ROHHAD have been described with the presence of neuroectodermal tumors (NET), now referred to as ROHHADNET (11,12). Hence, some believe that these two syndromes are part of the same fundamental disorder, perhaps sharing some common final pathway of clinical expression.

The clinical features of ROHHAD are phenotypically variable (Table 1).The phenotypic features common to all patients in the Ize-Ludlow’s study were alveolar hypoventilation and hyperphagia with the rapid onset of obesity. Acquired central hypoventilation has been described following trauma or tumor of the brainstem (13). The known occurrence of sudden death in ROHHAD, especially during sleep, necessitates caution in caring for these patients. It is not yet clear whether such deaths in ROHHAD occur because of cessation of respiration or because of other consequences of autonomic dysfunction such as primary hypotension, cardiac tachy, bradyarrhythmias.

Table 1. Phenotypes of 15 children with rapid-onset obesity, hypoventilation, hypothalamic dysfunction, and autonomic dysfunction (ROHHAD) [From (2)]

HD	n
Data are presented in the order of frequency of findings. SIADH indicates syndrome of inappropriate antidiuretic hormone secretion; IGF, insulin-like growth factor; IGFBP, insulin-like growth factor–binding protein.
Rapid-onset obesity	15
Failed growth hormone stimulation test	9
Hyperphagia	8
Polydipsia	8
Hypernatremia	7
Hyperprolactinemia	7
Diabetes insipidus	5
Hypothyroidism	5
Adrenal insufficiency	4
Hypodipsia	4
Polyuria	4
Short stature	3
Delayed puberty	2
Hyponatremia	2
Low IGF-1 and IGFBP-3 levels	2
Precocious puberty	2
Premature adrenarche	2
Transient SIADH	2
Amenorrhea	1
Hypogonadotropic hypogonadism	1
Irregular menses	1
Transient diabetes insipidus	1
Respiratory manifestations
Alveolar hypoventilation	15
Cardiorespiratory arrest	9
Reduced carbon dioxide ventilatory response	9
Obstructive sleep apnea	8
Cyanotic episodes	4
Developmental disorder
Developmental delay	3
Developmental regression	3
Autonomic dysregulation
Ophthalmologic manifestations	13
Thermal dysregulation	11
Gastrointestinal dysmotility	10
Altered perception of pain	8
Altered sweating	8
Cold hands and feet	6
Bradycardia	5
Tumor of neural crest origin	5
Syncopal episodes	1
Other findings
Abnormal brain MRI scans	7
Seizure	5
Enuresis	4
Hypotonia	4
Asthma	3
Hypercholesterolemia	3
Scoliosis	3
Hypersomnolence	2
Recurrent pneumonia before diagnosis	2
Deceased	1
Impaired glucose tolerance	1
Type 2 diabetes mellitus	1
Behavioral disorders
Depression	2
Flat affect	2
Psychosis	2
Behavioral outbursts	1
Bipolar disorder	1
Emotional lability	1
Obsessive-compulsive disorder	1
Oppositional-defiant disorder	1
Tourette’s syndrome	1
Hallucinations	1

Many interventions for stabilizing patients with dysautonomia have been described. Table 2 summarizes the current best-supported practices for patients with autonomic failure. We believe these are appropriately applied to children with ROHHAD in the perioperative period [Table from dysautonomia article, Anesthesiology 01/2012 (14)]. Interestingly, there are some data to suggest that relatives of patients with CCHS may also have dysautonomia (15).

Table 2. Summary of anesthesia implications [modified from (14)]

Preoperative factors

Patients with orthostatic hypotension with or without supine hypertension

Be aware of blood volume status for the conduct of anesthesia

Be aware of slight anemia owing to reduced erythropoiesis

Consider holding oral vasopressors to prevent drug interactions intraoperatively

Intraoperative factors

General vs regional anesthesia choice would be dependent on procedure type

Hypothermia and hyperthermia are both risks with poor thermal regulation

Patient positioning may have significant effects on vital signs

Nitroglycerine, prone positioning, or reverse Trendelenburg positioning can be used for supine hypertension

Sudden blood volume changes may precipitate untoward blood pressure response (rapid infusions or rapid blood loss)

Pneumoperitoneum should be applied with extreme caution or avoided if possible

Vasopressors and inotropes can have resistance or undue sensitivity

Abnormal electrolyte values affect responses to medications

Postoperative factors

Anticipate volume changes in the postoperative period and observe with a higher level of care

Orthostatic hypotension requires good blood volume and assistance with first ambulation

Abnormal electrolyte values affect responses to medications

Thermal regulation may be an issue in the recovery period

Ileus can be worsened by opiate use in the postoperative period

Restart home dysautonomia medications before discharge

Children with CCHS present earlier for procedures related to autonomic dysfunction. This includes GI procedures for Hirschsprung’s disease, which is more severe in these patients (16), gastrostomy tube placement, as well as intra-abdominal procedures, procedures related to abnormal ventilation (diaphragmatic pacing, tracheostomy placement), and other procedures for autonomic dysfunction (permanent pacemaker placement). ROHHAD patients, because of their later onset, often present for procedures related to alveolar hypoventilation. However, depending on the severity of dysautonomia, they may also present for any of the procedures described with CCHS as well as procedures required commonly for any child.

The nature of the hypoventilation in these syndromes can vary considerably. Many of the patients start out requiring only respiratory support at night and often progress to require additional daytime support. The variance in clinical course has led to these syndromes being placed in the rubric of Autonomic Nervous System Disorders (ANSD), which encompasses a variety of disorders that share the common feature of autonomic dysregulation (17).

Anesthetic management of these patients must take into account multiple factors. The diagnosis is often unknown at the time of presentation for surgery and may be found postoperatively when the patient cannot be weaned from ventilatory support (18,19). When the diagnosis of ROHHAD is known, a careful assessment for all areas of possible abnormality should be completed. These include gastroparesis with possible full stomach despite adequate fasting time and possible presence of a pacemaker. Other concerns include electrolyte abnormalities, seizure disorders, endocrine disorders, thermal dysregulation, and behavioral disorders. Awareness may allow the anesthesiologist to suspect the presence of ROHHAD in children with suggestive histories and examinations, even when the diagnosis has been neither entertained nor made.

Anecdotally, based on the previous use of premedication for our patients by other anesthesiologists, we believe judicious premedication is of benefit for anxiolysis in these children who have complicated lives, often have behavioral disorders, and require frequent medical interventions. Intraoperatively, the use of short-acting agents with minimal respiratory effects is appealing, although we note that in one of our cases, the use of sevoflurane had no measurable impact on the endtidal PCO2.

On the basis of history, our two patients have undergone a total of at least seven uneventful anesthetics, all with no problems. Despite our two cases with uneventful anesthetics, it appears prudent to assure careful postanesthetic respiratory, cardiac, and blood pressure monitoring until more is known about the manner in which children with ROHHAD react to anesthetic agents.

In conclusion, a new and growing spectrum of syndromes with potentially significant anesthetic implications has been discussed. We encourage other practitioners to report further experiences with children with ROHHAD and ROHHADNET to facilitate the development of guidelines for their anesthetic management.

Acknowledgments

This research was carried out without funding.

Conflict of interest

No conflicts of interest declared.