Over the past two decades, expanding legalization and changing public attitudes toward cannabis have led to a marked rise in its use across the United States. National estimates report up to 15.3% of the adult population used cannabis in the last year, and both the frequency and proportion of daily users have nearly doubled during this period. Among patients with gastroparesis cannabis use has similarly increased, with national estimates suggesting that roughly 1 in 10 affected individuals report current use [15–18]. Prior studies have linked cannabis use to greater symptom severity and altered healthcare utilization in the medically managed gastroparesis population, but the impact of cannabis on surgical outcomes has not been well characterized [15, 16, 19]. In this large, retrospective cohort study, we found that patients who used cannabis prior to surgical intervention for gastroparesis had a substantially higher symptom burden, particularly nausea, vomiting, abdominal pain symptoms, and were more likely to present with comorbid diabetes and GERD. Postoperatively, cannabis users experienced significantly higher rates of early surgical reintervention, greater short- and long-term inpatient utilization, and persistently lower outpatient follow-up. Despite these differences, long-term rates of reintervention and gastrectomy did not differ significantly, suggesting that preoperative cannabis use may be a surrogate for higher symptom burden and healthcare barriers, rather than diminished surgical efficacy.

In our surgical cohort, 7.25% of patients had documented preoperative cannabis use. This proportion is slightly lower than that reported in broader gastroparesis populations, where multicenter cohort data show 11.7% of patients reporting current use and national inpatient data demonstrate a 9% prevalence during hospitalizations for gastroparesis [18, 20]. The lower prevalence in our study may reflect undercoding in the preoperative setting, differences in patient selection for surgery, or patient reluctance to disclose use before an elective operation. Regardless of the cause, patients with gastroparesis who use cannabis represent a distinct subset of surgical candidates who should be identified preoperatively for accurate preoperative risk assessment and perioperative planning.

Cannabis users in our cohort exhibited a greater preoperative symptom burden, most prominently nausea, vomiting, and abdominal pain and greater rates of diabetes and GERD, Similar trends have been observed in prior studies of the impact of cannabis use in medically managed gastroparesis. One such study found that patients with gastroparesis who use cannabis reported significantly higher vomiting and upper abdominal pain scores compared to non-users (2.7 vs 2.1 and 3.5 vs 2.9, respectively). Additionally, cannabis users demonstrated greater use of antiemetics such as promethazine and cannabinoid-based therapies such as dronabinol [20]. Also consistent with our findings, large national database study demonstrated that 45% of gastroparetic cannabis users had diabetes compared to 30% of gastroparetic non-users (p < 0.001) [19]. These consisted data demonstrate that cannabis users constitute a more symptomatic and metabolically complex subset of the gastroparesis population which should be taken into account when they present seeking intervention.

The greater baseline symptom severity among cannabis users likely contributes to their higher postoperative healthcare utilization. Within 90 days of surgery, these patients had significantly higher rates of reintervention and hospital admission yet participated less in outpatient follow-up. This imbalance persisted over long-term follow-up, with cannabis users continuing to demonstrate higher inpatient use and lower outpatient engagement. Similar trends have been described in non-surgical populations. In a 1:1 propensity-matched cohort study of 41,347 patients with gastroparesis who did vs did not use cannabis, they found that cannabis use was associated with significantly higher odds of emergency department visits (OR 1.73) and hospitalizations (OR 1.44) [19]. Similarly, a National Readmission Database study identified cannabis use as an independent predictor of 30-day readmission following hospitalization for gastroparesis (OR 1.7) [21]. Our results extend these observations to the surgical population, suggesting that cannabis use may be a marker for a high-utilization phenotype even after definitive interventions such as pyloric drainage or gastric stimulator placement.

Interestingly, despite a higher rate of inpatient admissions, there is some evidence to suggest that when cannabis users are admitted, they have better inpatient outcomes. A National Inpatient Sample database study of nearly 1.5 million gastroparesis patients found that cannabis users hospitalized for gastroparesis had shorter lengths of stay (3.7 vs 5.5 days, p < 0.001) and lower in-hospital mortality (0.1% vs 0.9%, p < 0.001) compared to non-users [22]. However, these findings may reflect confounding from younger age, lower comorbidity burden, and lower socioeconomic status demonstrated in their study (all p < 0.001). Similarly, our cannabis-using patients were significantly younger, which may contribute to faster recovery and shorter hospitalizations. Future studies should clarify whether the higher admission rates observed among cannabis users stem from greater symptom burden or from socioeconomic barriers that limit outpatient access and drive reliance on emergency care.

The higher healthcare utilization observed among cannabis users in our study may reflect both incomplete symptom control and differences in how cannabis is used across patient populations. Many patients turn to cannabis to relieve refractory nausea, vomiting, or abdominal pain, yet its real-world effectiveness remains uncertain. In our surgical cohort, cannabis users presented with greater preoperative symptom burden and required more inpatient care postoperatively, suggesting that cannabis often provides inadequate or transient relief. Small therapeutic studies have reported symptomatic improvement with cannabinoid agents; for example, an open-label trial of 24 patients treated with marijuana, dronabinol, or both demonstrated significant reductions in GCSI nausea/vomiting, fullness/satiety, and bloating/distention scores [23]. However, these findings are limited by small sample size, lack of controls, and non-standardized dosing. The disparity between controlled studies and population-level outcomes likely reflects differences between medically supervised use and self-directed or recreational consumption, as well as the tendency for more symptomatic patients to self-medicate.

Beyond inadequate symptom control, physiologic and anesthetic impact from cannabis itself may also contribute to the higher early reintervention and hospitalization rates. Prior studies on non-gastroparetic cannabis users have demonstrated increased perioperative risk in this population. One large multicenter study reported that cannabis use disorder was associated with broadly increased in perioperative morbidity and mortality after any major elective non-cardiac surgery (OR 1.19) [24]. These results suggest that cannabis can affect perioperative outcomes independent of gastroparesis. Another study demonstrated altered anesthetic requirements, heightened postoperative pain, and more frequent nausea and vomiting among cannabis users [13]. Together, these findings reinforce that cannabis use may predispose surgical patients to greater postoperative care needs, which may contribute to early increased utilization and need for additional intervention. Conversely, a propensity-matched analysis of colectomy patients, found no significant association between preoperative cannabis use disorder and composite morbidity, anastomotic leak, or length of stay [25]. This difference likely reflects variation in disease pathology, surgical complexity, and postoperative physiology between colectomy and gastroparesis operations. In our cohort, the combination of greater preoperative symptom burden, cannabis-related physiologic effects, and distinct perioperative responses likely contributed to the higher rates of early postoperative utilization observed among cannabis users.

Prior research may have characterized cannabis use patterns, symptom severity, and healthcare utilization in gastroparesis broadly, yet there remains a critical gap in understanding its impact in the surgical setting. No published studies to date have examined outcomes following pyloric drainage or gastric stimulator placement in the context of cannabis use. Most available literature focuses on medical management or general hospitalizations, without attention to the unique perioperative and postoperative considerations of surgical gastroparesis care. Our findings therefore extend the existing body of knowledge by identifying cannabis use as a potential risk factor for increased early postoperative morbidity and long-term inpatient utilization in this specific population. Recognizing this association has important implications for preoperative counseling, perioperative planning, and postoperative follow-up strategies, and highlights the need for prospective studies to clarify causality and guide targeted interventions.

This is the first study to specifically evaluate the impact of cannabis use on postoperative outcomes in patients undergoing surgery for gastroparesis using a large, multi-institutional, real-world dataset. The findings are strengthened by both short- and long-term follow-up and the inclusion of surgical procedures relevant to advanced gastroparesis care. However, several limitations must be acknowledged, including the retrospective design and the limited granularity inherent to large administrative databases. Cannabis use was identified through ICD-10 coding, which likely underestimates true prevalence and does not capture usage patterns, dosage, or cannabinoid composition. Insurance status and other access-to-care variables were not available in the database and therefore could not be assessed as potential confounders. Unmeasured factors such as socioeconomic status, mental health comorbidities, and barriers to care may also influence both cannabis use and postoperative outcomes. Despite these limitations, the large and diverse study population provides broad generalizability, and the consistency of findings across multiple time points supports the robustness and clinical relevance of the observed associations. Procedure-specific subgroup comparisons were not performed, as the goal of this study was to evaluate cannabis use as the primary exposure across the surgical gastroparesis population rather than to compare surgical techniques.

Cannabis use appears to identify patients with greater clinical complexity rather than directly impairing surgical outcomes. Recognizing this association provides an opportunity to improve perioperative planning and patient engagement. Incorporating cannabis use into preoperative assessment can help tailor counseling, set realistic expectations, and coordinate resources for postoperative support and outpatient continuity. For gastroenterologists, awareness of this relationship may prompt earlier multidisciplinary involvement for patients whose symptom burden or healthcare utilization suggests elevated risk.

Future research should move beyond association to mechanism. Prospective studies must quantify cannabis exposure in terms of frequency, route, and composition, and evaluate its physiologic effects on gastric motility, pain response, and anesthetic requirements. Equally important is understanding patient behavior: why individuals with gastroparesis turn to cannabis, how it shapes their engagement with care, and whether barriers to outpatient access drive reliance on acute services. Clarifying these factors will enable development of targeted perioperative protocols and refine risk stratification for the growing subset of surgical patients who use cannabis.