Enhanced Recovery After Surgery for Noncolorectal Surgery?

A Systematic Review and Meta-analysis of Major Abdominal Surgery

Anthony Visioni, MD; Rupen Shah, MD; Emmanuel Gabriel, MD, PhD; Kristopher Attwood, PhD; Moshim Kukar, MD; Steven Nurkin, MD


Annals of Surgery. 2018;267(1):57-65. 

In This Article


This is the first meta-analysis of ERAS protocols across multiple elective abdominal surgical procedures. It is also one of the largest studies performed with regard to the number of studies and patients included. The primary outcome measure of LOS was significantly reduced in patients treated on an ERAS protocol. For all included studies, this reduction was 2.5 days (95% CI: 1.8–3.2). This represents not only a statistically significant result but a clinically meaningful one as well. This result was also consistent when only randomized trials were evaluated, with a mean LOS reduction of 2.6 days (95% CI: 1.7–3.5). In addition, this result is consistent with the Cochrane Review meta-analysis of colorectal surgery which showed a reduction in hospital stay of 2.9 days (95% CI: 2.2–3.7)[6] and another meta-analysis of colorectal cancer by Adamina et al[47] that showed a reduction in LOS of 2.5 days (95% CI: 1.1–3.9).

The other primary outcome measure of complications was significantly reduced when evaluating all studies, OR 0.7 (95% CI: 0.56–0.86). This, however, was not statistically significant when evaluating only randomized trials. Complications were defined by the authors of the original study and were variable depending on the type of procedure. These were often surgery specific (eg, pancreatic fistula rates in pancreaticoduodenectomy); however, they also included relatively minor complications (eg, vomiting). The complication rates noted in this study, therefore, represent patients who had any grade of complication postoperatively. It was not feasible in this study to perform a subgroup analysis of patients with major and minor complications given that the authors of the included studies do not use a uniform system of grading severity. However, none of the studies that evaluated mortality rates demonstrated a significant difference.

Regarding secondary outcome measures, ERAS protocol patients were noted to have a decreased time to first flatus of 0.8 days (95% CI: 0.4–1.1). Although this may seem like a modest reduction, it does demonstrate a functional improvement in patients treated with ERAS protocols. If all the benefit of an ERAS protocol was related to having the patient on a timed pathway, it is unlikely that a functional outcome such as time to first flatus would be significantly improved. Moreover, increasingly important in today's current healthcare environment, hospital costs were significantly decreased in patients on an ERAS protocol. Ten randomized studies included data on costs and showed a mean reduction of $5109.10 (95% CI: $4365.80–$5852.40) when compared with patients not on an ERAS protocol. Given that this study did not show a decrease in complications, it is likely that these savings were realized entirely from decreased LOS. Hospital cost research is notoriously difficult because of lack of standardization in actually determining these costs. However, even a modest decrease in hospital costs applied over multiple surgical disciplines has a strong possibility of significant savings.

Importantly, the decrease in LOS does not come at the price of increased hospital readmissions. Readmission rates for ERAS protocol patients were not statistically different from control patients in this study, OR 1.03 (95% CI: 0.84–1.26). This was based on evaluation of 25 randomized and cohort studies. However, the one study which did show a statistically significant increase in readmission rate was by Bu et al,[12] a randomized trial of gastrectomy in the elderly. This study showed that patients of age 75 to 89 years undergoing either distal or total gastrectomy had a readmission rate of 19% with ERAS protocol and 5% with conventional treatment (P = 0.013). Aguilar-Nascimento et al[25] attempted to answer whether ERAS protocol was safe in the elderly undergoing various abdominal surgeries. Although they did not report readmission rates, ERAS was safe in the elderly as complications were lower. Importantly, the median age of their ERAS cohort was 67.5 years. In the Bu et al study,[12] the average age was 80 years. This discrepancy may account for the differences in conclusions regarding the elderly. No definitive statements regarding the safety and efficacy of ERAS protocols in the elderly can be made from the current meta-analysis, and further research into this specific patient population is warranted.

As with any meta-analysis, publication bias is of particular concern. This was addressed in the current study through a search of ClinicalTrials.gov to determine if any negative studies were performed and not published. No studies were identified. Also funnel plots generally demonstrated low risk of publication bias in this meta-analysis. The exception to this is the secondary outcome measure of time to first flatus. As this was a secondary outcome measure, our study was not specifically designed to reduce the risk of publication bias in time to first flatus.

It should be noted that one study was identified in the systematic review but excluded because of being predominantly based on colorectal surgeries.[48] This study initiated 7 ERAS interventions and found no statistically significant reduction in complications or readmissions. The author also showed no difference in median LOS for ERAS protocol and conventional treatment, 6 days versus 6 days (P = 0.49), respectively. However, the authors state that only 3 of 7 of the interventions were successfully implemented and that compliance to these interventions "cannot be assumed." Lack of evaluation of compliance or successful implementation of ERAS protocols is a weakness of the current meta-analysis and many other studies within the field.

This meta-analysis also demonstrated a high degree of statistical heterogeneity in the primary end-points. The Q statistic ranges from 26.1–433.4 with P < 0.05 for all and the I2 statistic ranges from 43.2% to 95.6%. The statistical heterogeneity in this study most likely originates from the clinical heterogeneity of the included studies. This meta-analysis combined outcomes from various major abdominal operations. Although this serves to improve the generalizability of the results it does limit the inferences that can be drawn. We did attempt to account for this heterogeneity in our statistical analysis by using the random effects models as opposed to fixed effects models.

Another major critique of research into ERAS protocols is the low quality of the data, as noted in the Cochrane reviews. Most studies are cohort studies that lack direct comparability of patient groups or randomized trials that are not blinded. Inherent to both study designs is the inability to separate improvement in outcomes based on physiology from the Hawthorne effect. This is true for the studies that were included in this meta-analysis as well. It is unlikely that a truly blinded, randomized trial regarding ERAS protocols will be performed because of lack of feasibility. Also with a growing body of literature that suggests that ERAS protocols are safe, many centers are adopting these protocols without the expense and difficulty of performing a blinded, randomized trial. ERAS protocols are also expanding to specialties outside the scope of this study, including plastic surgery,[49] vascular surgery,[50] pediatrics,[51] kidney transplant,[52] and even selected emergency procedures.[53] It should also be noted that even if the outcomes for these studies are related to the Hawthorne effect and not necessarily to improving the surgical stress response, the decreased LOS without increasing complications or readmissions is in itself a worthy outcome that is being realized.

In addition to decreased LOS, there is evidence of other potential benefits to ERAS protocols. As hospital systems incorporate patient satisfaction scores into quality assessments, interventions to improve patient outcomes will by necessity evaluate if the patients are satisfied with these interventions. Although data on this subject is limited, Khan et al[54] performed a systematic review that demonstrated that health-related quality of life and patient satisfaction scores were not adversely affected by treatment in an ERAS program. Interestingly, patient satisfaction is not decreased with these pathways even with a concomitant decrease in opioid usage. Avoidance of opioids is one of the tenants of ERAS protocols, and has even garnered the attention of the US Surgeon General in a letter to healthcare providers.[55] Patients treated within these programs can have improved pain control with less opioid analgesia.[56]

Finally, for patients undergoing surgeries as part of multimodal cancer treatment plan, a faster recovery that allows the patient to initiate adjuvant treatments can potentially improve their oncologic outcomes.[57] A study by Day et al[58] found on multivariate analysis that patients undergoing liver surgery were more likely to return to baseline functional status if they were treated according to an ERAS protocol (OR 2.62, 95% CI: 1.15–5.94). This return to baseline status correlated with a trend toward an increased percentage of patients getting to their intended postoperative oncologic therapy (95% vs 87%) and in less time (44.7 days vs 60.2 days).

In conclusion, this meta-analysis demonstrates a decreased LOS and cost by not increasing complications or readmission rates in noncolorectal surgical patients undergoing abdominal operations when ERAS protocols were implemented to their perioperative care. It adds to the evidence that ERAS protocols do not increase complication or readmission rates and are beneficial to surgical patients and the healthcare system overall. It is time to consider ERAS pathways as a method to expedite the return to normal physiology after the controlled trauma of surgery, a principle that consistently works across procedures and disciplines. With continued improvements in outcomes, ERAS protocols will likely become standard practice for major abdominal surgeries, and will be a major component to improving quality and value of surgical care.