Introduction

Enhanced recovery after surgery (ERAS) protocols have been shown to improve patient outcomes and decrease the length of stay (LOS) without an increase in morbidity [1, 2]. Specifically, after bariatric surgery, ERAS protocols have demonstrated improved post-operative pain and nausea control, and a reduction in LOS [1].

The opioid epidemic is a major public health crisis in the USA, and opioid overdose results in significant morbidity and mortality [3]. With the widespread adoption of ERAS protocols, one of the primary modes of pain control is the use of opioids. To mitigate the harmful consequences of opioid overuse in the bariatric population, the metabolic and bariatric surgery accreditation and quality improvement program (MBSAQIP) introduced the bariatric surgery targeting opioid prescriptions (BSTOP) protocol [4]. The major reason for its creation was evidence showing that new-onset opioid dependence in post-operative bariatric patients accounts for more deaths than all other complications of bariatric surgery combined [4].

Our MBSAQIP Bariatric Centers of Excellence implemented both bariatric ERAS and BSTOP protocols at approximately the same time. The purpose of this study was to evaluate the combined impact of ERAS and BSTOP protocols on outcomes of bariatric surgery and the need for opioids.

Methods

This is a retrospective review of our prospectively maintained database for patients who underwent either a sleeve gastrectomy or Roux-en-Y gastric bypass (RYGB) at our MBSAQIP Bariatric Center of Excellence in New York. Pre-intervention data were from August 2018 to July 2019, while post-intervention data were from August 2019 to October 2020. The new combined ERAS and BSTOP protocol is detailed in Fig. 1. Criteria for discharge at our center are toleration of clear liquid diet without nausea or vomiting, adequate oral intake to endure oral hydration at home, and adequate pain control. The comparison post-intervention period was longer due to the COVID-19 pandemic, which led to no elective surgeries being performed for a few months. Patients who underwent revisional bariatric surgery or concurrent hiatal hernia were not excluded from analyses.

Fig. 1
figure 1

Enhanced Recovery After Surgery combined with Bariatric Surgery Targeting Opioid Prescriptions protocol

Full size image

The database was queried for information on demographic variables (age, gender, and race), comorbid conditions (obstructive sleep apnea [OSA], diabetes mellitus [DM], hypertension [HTN] and gastroesophageal reflux disease [GERD]), and anthropometric measures (weight, height; body mass index [BMI] was calculated). Data on post-operative emergency department visits and reoperations were retrieved. Primary outcomes of interest included length of stay, 30-day readmission, 30-day complications, and the proportion of patients discharged on opioids.

BSTOP data were collected by reviewing discharge medications and identifying patients discharged on opioids. Of note, the final implementation of BSTOP protocol was in October 2019 due to delay resulting from the COVID-19 pandemic. The proportion of patients discharged on opioids was compared between the pre- and post-intervention groups.

Demographic characteristics were summarized by pre- and post-intervention groups and presented as median (interquartile range [IQR]) or frequency (percentage), as appropriate. Baseline characteristics were compared between the groups using the Wilcoxon rank-sum test for continuous variables, and Chi-square or Fisher’s exact test for categorical variables. Multivariate Poisson regression (adjusted for age, sex, BMI, race, hypertension, diabetes, GERD, OSA, and OA) with robust standard error was used to analyze LOS data (in days). Deviance statistic and Akaike information criterion (AIC) were used to assess the goodness-of-fit for this model.

Our combined ERAS and BSTOP protocols are detailed in supplemental data. SAS 9.4 was used to analyze all data. Statistical significance was assumed if p value < 0.05.

Results

Our data showed that the number of patients that underwent bariatric surgery in the pre- and post-intervention period were 360 and 297, respectively. Overall, the mean (range) age was 45 (36–54) years in study participants. In pre- and post-intervention groups, mean age was 44 years and 47 years respectively. Seventy-eight percent of all patients were female. The majority of patients (overall 58.8%; 57.9% and 59.9% in pre- and post-intervention groups, respectively) were Caucasians. The most common comorbid condition was GERD (68.0%), followed by OSA (59.7%). There was no significant difference between the groups in demographic characteristics, comorbid conditions, or preoperative BMI (p value > 0.05 for all comparisons), see Table 1.

Table 1 Demographics and clinical characteristics
Full size table

In unadjusted analyses, there was no significant difference in 30-day emergency room visits (pre-intervention 3.3% [n = 12/360] vs. post-intervention 4.0% [n = 12/297]; p value = 0.631), 30-day readmission rate (pre-intervention 4.4% [n = 16/360] vs. post-intervention 5.4% [n = 15/297]; p value = 0.577) or 30-day complication rate (pre-intervention 4.2% [n = 15/360] vs. post-intervention 6.4% [n = 19/297]; p value = 0.199). LOS was significantly lower in the post-intervention group; mean (IQR) 2 (1–2) days vs. 1 (1–2) day, p value < 0.001. Table 2 details results for 30-day outcomes.

Table 2 Unadjusted outcomes for pre and post intervention groups
Full size table

After adjusting for all other factors, patients in the post-intervention group were expected to have 0.74 (95% CI 0.65–0.85; p value < 0.001) times lower LOS compared to patients in the pre-intervention group. Overall, patients with DM had a significantly longer LOS (RR: 1.22; p = 0.018). No other covariates were associated with LOS (p value < 0.05 for all) (Table 3).

Table 3 Multivariable Poisson regression model for length of stay
Full size table

BSTOP analysis found a significant difference between the two groups. In the pre-intervention group, 40.6% of patients (n = 146/360) were discharged on opioids, which decreased to 7.1% post-intervention (n = 17/239).

Discussion

Our results show that concomitant introduction and implementation of ERAS and BSTOP protocols results in reduced LOS without an increase in readmission or complication rates. Furthermore, it also results in a significant reduction in need for opioid medications at discharge.

Interestingly, our results showed a significantly longer length of stay in diabetic patients post-operatively. It is difficult to pinpoint exactly the reason for this finding. However, we believe this might be a result of residual confounding from lifestyle factors associated with diabetes, such as a sedentary lifestyle, decreased physiological reserve, poor exercise tolerance slowing down post-operative recovery, and return to baseline physical status. Other factors contributing to residual confounding may include unadjusted comorbid conditions such as depression and anxiety. Further, diabetes-associated gastroparesis and gut dysmotility may also contribute to poor post-operative oral intake and delayed discharge.

The effectiveness of ERAS protocols in decreasing LOS without an increased risk of complications or readmission is well-documented [1]. Data from 310 MBSAQIP accredited centers show that BSTOP protocols are effective in reducing the need for opioids both in inpatient setting and at discharge [5, 6]. However, this is the first study investigating the introduction of ERAS and BSTOP at the same time, and shows a major reduction in opioid use (7.1% vs. 40.6%) without an increase in complications. Of the small proportion of patients discharged on opioids post-intervention (7.1%), it is possible that a significant proportion of them had pre-existing conditions that predisposed them to poor postoperative pain control, necessitating the prescription of opioids at discharge.

The advantages of reducing opioid use in bariatric patients are well-documented. Opioid-induced respiratory depression in the setting of OSA, a condition common in the bariatric population, is especially associated with worse outcomes [7]. Importantly, data from the longitudinal assessment of bariatric surgery-2 cohort showed that in patients with no reported preoperative opioid prescriptions, 5.8% and 14.2% had regular prescribed opioid use at six months and seven years postoperatively, respectively [3], suggesting that bariatric patients are particularly prone to developing opioid dependence. Moreover, opioid dependence is the single most common cause of mortality in bariatric surgery patients and causes more deaths than all other bariatric surgery complications combined [4]. This highlights the need for evidence-based interventions to limit opioid use and dependence in the bariatric population. Conversely, inadequate postoperative pain management is associated with increased risk for chronic pain and thus opioid use, which underscores the need for a multimodal pain regimen to ensure adequate pain control, and use of opioids if necessary [8, 9].

Protocols such as BSTOP are also important from a health systems standpoint. Globally, around 580,000 individuals undergo bariatric surgery annually, a major proportion of which, 198,651 people in 2020, occurs in the USA [10, 11]. The Center for Disease Control and Prevention (CDC) estimates the total cost of prescription opioid misuse in the United States to be $78.5 billion annually. Therefore, a reduction in opioid dependence in post-bariatric surgery patients may also have significant economic implications for the health system, especially in the USA [12].

Our protocols are strictly evidence-based and designed based on previously published guidelines [13]. Patients undergo extensive pre-operative education, and expectations are set for all phases of care. Multimodal pain and nausea control begin in the preoperative phase. In the intraoperative phase, opioid-free anesthesia, which is associated with a reduced risk of postoperative nausea and vomiting [14], is given, including on induction. Anesthetic gasses are also avoided due to their association with increased postoperative nausea [15]. Transversus abdominis plane (TAP) block, which has been shown to significantly decrease postoperative pain, and lower opioid use, nausea and vomiting, is performed on all patients under laparoscopic guidance [16,17,18]. Additionally, TAP block also promotes early ambulation and improves patient satisfaction [18, 19]. Post-operatively, multimodal pain and nausea control are continued while in the hospital and on discharge.

Importantly, part of our protocol includes only dosing breakthrough intravenous opioids after the patient has been evaluated by a provider. This aspect may be more difficult to implement in hospitals without in-house providers present 24 h a day. However, with proper education of patients and nurses, hospitals with this limitation may be able to overcome that hurdle.

Our study has several limitations. It does not investigate patient satisfaction or opioid use on long-term follow-up. Additionally, our BSTOP protocol was implemented in October 2019, 2 months after the introduction of ERAS; however, we believe that the huge difference in opioid use between the pre- and post-intervention periods still demonstrates the effectiveness of combined implementation of both protocols. Finally, a subgroup analysis investigating the combined impact of these protocols in sleeve gastrectomy and RYGB groups was not performed due to power considerations.

Conclusion

ERAS and BSTOP protocols reduced length stay and opioid need at discharge without an increase in complication or readmission rates. Their impact on other foregut surgery procedures should be evaluated.