The response rate to prophylactic TCA in our study (46/75, 61.3%) is broadly consistent with the 70% response rate reported in the literature [16, 18, 22, 29, 30]. Intolerance to TCA therapy was only reported in three (4%) patients, which is significantly lower than the 26%–34% intolerance rate reported at mean doses of 75–100 mg [17, 18]. These results from a single center in the UK suggest that the therapeutic effect of TCA may be achieved at a lower dosage than that prescribed in North America (75–100 mg daily) with a potentially more favorable side‐effect profile [16]. The response rate observed at a lower dosage among patients with CVS in the UK compared to those in North America might be explained by differences in symptom severity and/or genetic factors that influence neuromodulator metabolism [29].

Seventeen (22.7%) patients in our study reported heartburn, consistent with previous research demonstrating a 19% prevalence of gastroesophageal reflux disease among individuals with CVS [31]. In the multivariate logistic regression model, heartburn was independently associated with a lack of response to the treatment algorithm. It is tempting to hypothesize that patients with concomitant heartburn have a greater degree of visceral hypersensitivity, suggestive of a more dysregulated brain–gut axis, which would necessitate a potentially higher TCA dosage or combination therapy with nonpharmacological agents.

Perhaps unsurprisingly, an anti‐emetic prescription was associated with an increased odds of response to the treatment algorithm. Although this did not reach statistical significance in the multivariate logistic regression model (p = 0.07), this finding still implies that anti‐emetics positively influenced the effectiveness of the treatment algorithm. Therefore, whether this treatment algorithm reduces patients' consumption of abortive agents (i.e., anti‐emetics) should be evaluated in future studies.

Thirty‐three percent of patients in our study were taking opioids, consistent with prevalence rates of 23% [17] and 28.6% [19] documented in the literature. Opioids are associated with nausea and vomiting [32], so it is perhaps not surprising that opioid consumption decreased the odds of clinical response to the treatment algorithm in the multivariate logistic regression model—the lack of statistical significance may be explained by underpowering. A previous study suggests that coexisting opioid use may be associated with lack of response to TCA therapy in the prophylactic management of CVS [22].

Nineteen percent of patients with CVS in our study consumed cannabis. The higher prevalence rate of cannabis use in North American settings (37% [33] and 39% [29]) may be explained by the fact that cannabis remains illegal for recreational use in the UK, unlike in some regions of the United States. Since recreational cannabis consumption is illegal in the UK, it is possible that some patients may not have been forthcoming about usage and that the prevalence rate could have been higher had urine toxicology been performed to confirm consumption. In the only UK study of CVS (n = 17), cannabis use was reported in five (29%) individuals [34]. In our study, cannabis use did not impact clinical response to the treatment algorithm on multivariate analysis (p = 0.3). Among the patients (n = 14) who reported current cannabis use, all self‐reported adherence to cannabis cessation advice, among whom six (42.9%) were in clinical remission at follow‐up. Other centers have reported less success with cannabis cessation—in one case series, 70% of patients with CVS adhered to cannabis cessation recommendations, among whom 86% were in remission [35].

In our center, patients who do not achieve a clinical response to the treatment algorithm despite cannabis and/or opioid cessation are routinely consented to undergo urine drug screening. In two such cases, patients were identified to have a positive urine screening for cocaine. This finding does not imply that clinicians should perform urine toxicology on all patients with CVS. Instead, it suggests that urine toxicology may be warranted in select cases, particularly those involving opioid dependence or the presence of other risk factors for illicit drug use [36].

Strengths of this study relate to its large sample size (n = 75) of well‐characterized patients with Rome IV CVS, which builds upon the findings of a case series conducted in the UK [34]. The relatively prolonged follow‐up (29.4 [20.2] months) also accounts for the episodic nature of CVS, allowing for the consideration of symptom variation that may occur during specific times of the year (e.g., secondary to psychosocial stressors).

Our study is not without limitations, particularly those intrinsic to any retrospective analysis which employs chart review and uncontrolled medication usage. Since data were collected in routine clinical practice, patients did not complete validated GI symptom questionnaires, so the degree of change in the frequency and/or intensity of GI symptoms cannot be quantified with absolute accuracy. To address this concern, two independent clinicians (CS and FC) reviewed patient charts, and discrepancies were resolved by the first (MFB) and senior author (MC). Given patients were drawn from one consultant in a single‐center setting, the characteristics of our sample may not be entirely representative of the CVS population on a national level. However, although patients in our study were consulted in a tertiary referral center, the majority (91.3%) of patients who responded to the treatment algorithm required first‐line treatment alone, which suggests that the severity of CVS may parallel that seen in non‐referral settings.

In addition to using validated questionnaires, future studies would benefit from studying the original indication for the opioid prescription. Whether this treatment algorithm reduces reliance on the number of abortive therapies (i.e., anti‐emetics), the frequency of CVS episodes, and the duration of the emetic phase is also worthy of further evaluation. Strategies to optimize continued outpatient engagement in treatment should be explored in the future. Indeed, non‐responders may have benefited from a higher TCA dosage, but from clinical experience, many in this group had infrequently engaged with the outpatient treatment plan, which likely accounts for the shorter follow‐up duration among non‐responders versus responders (36.1 vs. 24.8 months). Finally, compared to TCAs, there is a relative paucity of data addressing SNRIs, SSRIs, and topiramate in the prophylactic management of nausea and vomiting disorders [16, 25], so this single‐center study should encourage clinicians to integrate non‐TCA alternatives in treatment algorithms.

In conclusion, prophylactic low‐dose TCA, along with opioid and cannabis cessation, may be important strategies in the management of CVS.