Earlier clinical diagnosis has been facilitated by the recently defined consensus diagnostic criteria by NASPGHAN (2008) and Rome III (2006) criteria, the former being quantitatively more rigorous, i.e., requiring 3–5 vs. 2 total episodes [2] (Table 39.2). There is common confusion over the nomenclature as the older classification is “abdominal migraine” and the newer term since the 1990s is “cyclic vomiting syndrome” or “cyclical vomiting syndrome” (UK). Today, from an operational standpoint, the predominant and most consistent symptom during episodes defines the illness, i.e., abdominal pain is termed abdominal migraine, and conversely vomiting is denoted CVS. However, there is considerable clinical overlap because ~50 % of those diagnosed with abdominal migraine also vomit, and 80 % of those with CVS also have abdominal pain.

The continuum between CVS and migraine was suggested by Whitney in 1898 and corroborated by other authors including us in 1998 [3, 4]. In a cross-sectional school survey in Scotland, Abu-Arafeh described a developmental progression from CVS to abdominal migraine and migraine headaches, median ages 5, 9, and 11 years with prevalences of 1.9 %, 4 %, and 11 %, respectively [5]. This suggests a natural history that begins with CVS and ends with migraines. Although some experience all three phases, the largest group trades CVS for migraines by age 10. We estimate 75 % will develop migraine headaches by age 18 years (Li, unpublished data).

The previous lack of a specific ICD 9 code has rendered it difficult to establish the true prevalence of CVS. However, ICD 10 now includes a specific code (G43.A0) specific for CVS [6]. Typical misdiagnoses, including gastroenteritis, gastroesophageal reflux, food poisoning, and eating disorders, often delay accurate diagnosis by a median 2.5 years [7, 8]. At our GI clinic, CVS was second only to gastroesophageal reflux as a cause of recurrent vomiting [9]. Two school-based surveys estimated the frequency to be 2 % in Scottish and Turkish children (Table 39.3) [5, 10], and the incidence of new cases of CVS was reported to be 3.15 per 100,000 children per year in Irish children. In our series, the average age of onset of CVS is 4.8 years with predominance in girls over boys (57:43). Similar data was replicated in a large study from Iran [8].

CVS has a significant deleterious impact on the quality of life in affected children. Although well in between episodes approximately 90 % of the time, 58 % of affected children require intravenous fluids during episodes and average ten visits to the emergency department. School-age children miss an average of 24 days of school per year [7, 11]. Medical morbidity is reflected by the high average annualized cost of management of $17,000 in 1998 that includes doctor visits, emergency department visits, inpatient hospitalizations, missed work by parents, and biochemical, radiographic, and endoscopic testing [12]. A growing number of comorbid conditions such as anxiety and postural tachycardia syndrome (POTS) also contribute to functional disability. We have documented lower global quality of life scores than in healthy controls and those with functional GI disorders (irritable bowel syndrome) and equivalent to organic GI diseases (e.g., inflammatory bowel disease, gastritis, fatty liver disease) [13]. Nearly half (47 % overall, 59 % of school age children) of CVS sufferers meet criteria for an anxiety disorder and we found that anxiety was the prime predictor of impaired quality of life, even more than the quantitative severity of episodes [14].

In the absence of a defined etiopathogenesis, CVS remains classified as an idiopathic disorder. Recent investigations support the contributory roles of mitochondrial DNA (mtDNA) mutations and dysfunction, heightened hypothalamic–pituitary–adrenal (HPA) axis activation, and autonomic nervous system (ANS) dysfunction. CVS is a functional brain–gut disorder perhaps mediated through altered brainstem modulation of effector signals.

The evolving clinical picture of CVS has included an increasing number of associated comorbidities. In Boles’ series, 25 % had coexistent neurological findings of developmental delay, seizures, hypotonia, and skeletal myopathy as well as cognitive and cranial nerve dysfunction [33]. These children classified as CVS+ were found to have an earlier age of onset for CVS and a three- to eightfold higher prevalence of dysautonomic (neurovascular dystrophy) and constitutional (growth retardation) manifestations than CVS patients without neurological findings. Other comorbidities in non-neurologically impaired children include anxiety (47 %) and depression (14 %) [34], irritable bowel syndrome (67 %) [35], GERD (39 %), colonic dysmotility (20 %) [33], limited stamina or chronic fatigue (52 %) (Li, unpublished), sleep disturbance (onset or maintenance) (48 %) (Leung, Li, unpublished), POTS (19 %) (Leung, Li, unpublished), and complex regional pain syndrome (12 %) [36]. Often, these contribute to the poor quality of life and have to be treated concomitantly to help restore the child to functionality.

There appear to be subphenotypes of CVS, some of which overlap and may be present in the same patient. The 83 % that are migraine related (positive family or personal history) tend to have significantly less severe episodes that are more responsive to antimigraine therapy [37]. It now appears that the majority has a matrilineal inheritance pattern (for migraine and other functional disorders) and may have mtDNA single nucleotide polymorphisms and mitochondrial dysfunction [16]. Many appear to have predominantly sympathetic overtone and comorbid POTS in whom treatment of POTS helps reduce frequency of vomiting episodes. The Sato variant is associated with hypertension during episodes and an endocrine profile of heightened HPA axis activation. Those with long-interval calendar-timed episodes every 60+ days apart appear particularly difficult to treat. Boles has described a group with neurodevelopmental deficits in whom CVS begins early in life [33]. There are post-menarcheal girls with catamenial CVS who respond to low-estrogen birth control pills or ablation of menses.

A group of young adult males (>100 case reports) who use large amounts of recreational or medical marijuana over several years may in fact trigger CVS symptoms that have been labeled as cannabis-induced hyperemesis. However, it is more likely cannabis-triggered CVS [38]. Several series reports termination of bouts of emesis after cessation of chronic use of marijuana. Another case series and a large, anonymous survey of CVS indicate that marijuana users experience reduction in nausea and anxiety raising the possibility that marijuana may aggravate symptoms in some and mitigate them in others [39].

At present , there are no specific tests to diagnose CVS, and the diagnosis rests primarily upon fulfilling clinical criteria. The first step requires differentiating a cyclic or sporadic pattern (high intensity, low frequency) of vomiting in which extraintestinal disorders including CVS are most common from a chronic vomiting (low intensity, high frequency, e.g., daily) one in which upper GI tract disorders predominate [9]. Approximately 90 % of children who fulfill the NASPGHAN consensus criteria (Table 39.2) are ultimately found to have CVS [2, 9]. Most of the testing in undiagnosed children who present with recurrent vomiting is directed toward identifying underlying gastrointestinal, neurologic, renal, metabolic, and endocrine causes that can be discovered in the remaining 10 %. The challenge to the clinician is to determine which and how much testing should be performed, as the traditional “shotgun” approach is costly, time-consuming, and invasive.

The recent NASPGHAN Consensus Statement (2008) guidelines recommend against the traditional shotgun evaluation and for initial an upper gastrointestinal series to exclude malrotation and anatomic obstructions and a basic metabolic profile (electrolytes, glucose, BUN, creatinine) [2]. Further testing beyond that should be based upon specific warning signs (Table 39.4). In those who present with bilious vomiting and abdominal tenderness, abdominal imaging should be performed to exclude hydronephrosis, pancreatitis, and cholecystitis. In those in whom episodes are triggered by intercurrent illnesses, fasting, or high-protein meals, screening should be performed for urea cycle, fatty acid oxidation, disorders of organic and amino acid metabolism, and mitochondrial disorders. This screening has a better diagnostic yield in the early part of an episode of CVS before intravenous glucose and fluids are administered. Those presenting with abnormal neurological findings including altered mental status, papilledema, ataxia, or seizure should have a neurological evaluation and brain MRI considered. Presentation of CVS under the age of 2 should also prompt further metabolic or neurological testing [2].