Cyclic vomiting syndrome: a retrospective chart review
MetadataShow full item record
Purpose: We aim to characterize a large cohort of CVS patients seen at MGH in order to better understand this disorder. In addition, as CVS patients are known to have a higher cannabis use than those with other functional disorders, characteristics specific to CVS marijuana user patients have yet to be determined. Therefore, we aim to determine the variables that are patient specific predictors of cannabis use in CVS. Methods: All patients with a CVS diagnosis were seen at our facility as inpatient, outpatient or in the emergency department and medical record numbers were identified via Research Patient Data Registry (RPDR) query search tool. From the medical records, we verified a diagnosis of CVS based on Rome III criteria and collected information on demographics, co-morbidities, health care utilization and substance abuse history over a 16 year time period (1997-2013). We then proposed patient specific predictor variables of marijuana use based on our experience with CVS patients and incorporated these variables into a model for predicting marijuana use. We used this model to examine the effect of patient characteristics on marijuana use via logistic regression with estimation of odds ratio and 95% confidence ratio. Results: A total of 91 CVS patients were obtained, 67% of which were male with a mean age of 28 years old, average age of first attack of 20.2 years and employment or full time student status in 62% of patients. Psychiatric conditions were present in 73% of individuals with anxiety present in 50% and depression in 47%. The presence of a chronic pain syndrome was found in 40%. Marijuana use was present in 76% of our cohort, daily alcohol use 53%, and narcotic use 27%. Men with CVS were at significantly greater risk for marijuana use compared to women (OR .23, 95% CI .07-.77) as well as daily alcohol use (OR 5.26, 95% CI 1.469-18.828). Individuals with a chronic pain syndrome were at significantly lower risk for cannabis use (OR .15, 95% CI .039-.575) and psychiatric illness, age and narcotic use were found not to be associated with marijuana use. On average, individuals presented to the ED 1.6 times/year with one patient having a high of 208 ED visits/year. Of those that presented more than once during the time period in study there was a median interval between visits of 103.6 days. Conclusion: We found that CVS patients had significant psychiatric co-morbidities, chronic pain syndromes, and history of substance abuse. We found male gender and alcohol use to be two variables that were predictors of marijuana use in the CVS population while the presence of a chronic pain syndrome decreased the likelihood of marijuana use. The significant employment rate and full time student status of our cohort suggests a higher degree of functionality. Further prospective studies are needed to determine the role of marijuana use in the CVS population in terms of quality of life, health care utilization, and severity of disease.