Introduction

Enhanced recovery after surgery (ERAS®) is an evidence-based multimodal approach to optimize perioperative pathway which allowed to reduce postoperative morbidity and shortened length of hospital stay. Therefore, enhanced recovery protocols have been proposed as the standard of care in elective colorectal surgery [1, 2].

Despite the publication of proper guidelines [3], the spread of enhanced recovery protocols in patients undergoing emergency surgery is still limited. Preliminary results have been published in patients who underwent surgery for obstructing colorectal cancer or perforated peptic ulcers [4,5,6,7,8]. However, systematic reviews and meta-analysis showed that enhanced recovery protocols were different across the studies, especially concerning the intraoperative items [9, 10]. Limiting factors to the widespread of enhanced recovery protocols in emergency surgery are the non-applicability of preoperative items and the presence of acute stressful diseases often requiring tailored care rather than standardized protocols [11, 12].

The aim of this study was to evaluate the feasibility of an enhanced recovery protocol in a large cohort of patients undergoing emergency general surgery and to identify possible factors impacting postoperative protocol compliance.

Methods

This is a prospective, observational, multicenter cohort study promoted by the Italian Society of Emergency Surgery and Trauma and the Perioperative Italian Society. Eight Italian high-volume hospitals with previous experience in enhanced recovery protocols in major elective surgery were involved. The study protocol was shared during a multidisciplinary meeting involving surgeons and anesthesiologists from each center. Supplementary Table reports the study protocol which was approved by the Ethical Committee of the promoting center (n. 0,012,747 08/10/2020) and was registered on clinicaltrial.gov (identifier NCT04648644).

Patients aged   >  18 years undergoing unscheduled intestinal resections with or without anastomosis, hollow viscus injury repair, enteric bypass or adhesiolysis in presence of either peritoneal contamination or intestinal obstruction were included in the study. Exclusion criteria were refused to participate, septic shock, and emergency surgery for complications following elective surgery, operative endoscopy or diagnostic procedures. Patients presenting with multiple organ failure who required damage control surgery with open abdomen and/or immediate postoperative ICU stay longer than 72 h were dropped out from the study because they had no chance to adhere to the enhanced recovery protocol.

Demographics, body mass index, Charlson Comorbidity Index, primary diagnosis, type of surgery, adherence to each ERAS item and short-term outcome parameters were anonymously collected from all patients. Major complications have been classified according to the Clavien-Dindo scale [13]. Patients’ follow-up was carried out by office visits or telephone calls.

The primary endpoint of the study was the adherence to the enhanced recovery protocol. Secondary endpoints are 30-day mortality and morbidity rates, and length of hospital stay.

Statistical analysis

Continuous variables were described as median and interquartile range (IQR). Categorical variables were described as percentages. The cumulative adherence was recorded daily for each postoperative item. Univariate and multiple ordinal regression models were calculated using the number of achieved postoperative items as outcome. For each postoperative item, patients’ compliance was defined using the median value as the threshold. Variables with a significant association at the univariate analysis (p < 0.05) were adopted in the multiple ordinal regression model. Pearson’s linear correlation was calculated between the number of achieved postoperative items and length of stay. Statistical analysis was performed using the IBM SPSS Statistics 27 software (IBM Corp. Released 2020. IBM SPSS Statistics for Windows, Version 27.0. Armonk, NY: IBM Corp).

Results

Between November 2020 and November 2021 among the eight participating centers 909 patients fulfilled the inclusion criteria. Figure 1 reports the flow diagram of the study. A total number of 589 patients were included in the analysis.

Fig. 1
figure 1

Study flow diagram

Full size image

Table 1 shows that 242 (41.1%) patients were over 75 yrs., 318 (54.4%) had ASA 3–4, while body temperature was below 36.0 °C in 177 (33.5%) patients. Three hundred and twenty-seven (55.5%) patients had intestinal obstruction, while 262 (44.6%) had intra-abdominal infections. The median time from hospital admission to operation was 8 h (3–25). Two hundred and fifty-six (43.5%) patients underwent intestinal resection with anastomosis and 93 (15.8%) underwent intestinal resection without anastomosis. One hundred and forty-four (24.6%) patients were treated with minimally invasive techniques.

Table 1 Patients’ characteristics
Full size table

Table 2 reports adherence to the enhanced recovery items. The highest adherence was obtained for operative warming and postoperative nausea and vomiting prophylaxis. The median operative fluid infusion was 12 (8.33–17.14) mL/Kg/h. Drains were placed in 92.7% of patients with intra-abdominal sepsis and in 55.7% of patients with intestinal obstruction.

Table 2 Adherence to enhanced recovery items
Full size table

Table 3 summarizes the postoperative short-term outcome. Overall morbidity was 60.4%, major complications occurred in 92 (15.6%) patients, and 30-day mortality was 6.3%. Median length of hospital stay was 8 (6–12) days.

Table 3 Short-term postoperative outcome
Full size table

Table 4 shows the cumulative compliance to postoperative items. The median adherence was reached on postoperative day (POD) 1 for naso-gastric tube removal (55.3%), on POD 2 for mobilization (68.8%) and urinary catheter removal (57%), and on POD 3 for oral intake (56.4%) i.v. fluid stop (52.3%).

Table 4 Cumulative compliance to postoperative items
Full size table

Table 5 shows that laparoscopy was positively associated with an increasing postoperative compliance at the ordinal logistic regression analysis. A negative association with postoperative compliance was found for preoperative hyperglycemia, operative fluid infusion > 12 mL/Kg/h, presence of abdominal drain, perforated peptic ulcer repair, duration of surgery and major complications. There was a linear correlation between the increasing postoperative items compliance and the length of hospital stay (r = −0.552, p < 0.001) as shown in Fig. 2.

Table 5 Univariate and multiple ordinal regression analyses for postoperative recovery items
Full size table
Fig. 2
figure 2

Linear correlation between number of postoperative items achieved and length of hospital stay

Full size image

Discussion

The present study supports the implementation of enhanced recovery protocols in patients undergoing emergency gastrointestinal surgery. The median adherence to postoperative recovery outcomes was reached on postoperative day 1 for naso-gastric tube removal, on day 2 for mobilization and urinary catheter removal and on day 3 for oral feeding and i.v. fluid suspension. Laparoscopy was associated with an earlier recovery, whereas preoperative hyperglycemia, fluid overload, abdominal drain, duration of surgery, and major morbidity were associated to a delayed recovery.

Few studies have been published on patients who underwent emergency surgery, the majority focused on patients with obstructing colorectal cancer or perforated peptic ulcers [4,5,6,7,8]. Enhanced recovery protocols were associated with shorter length of stay and lower postoperative complications when compared to traditional care. Roulin and coll. reported a lower compliance to postoperative protocol in patients who underwent urgent colectomy when compared to patients who underwent elective colectomy [6].

The present study enrolled consecutive patients without any selection bias. In fact, the majority of patients were elderly with ASA score >  2 and the median time from admission to surgery was short, thus reflecting the daily practice of emergency surgery. Patients’ compliance to postoperative items was satisfactory with all targets reached one-day later in comparison with what has been reported following elective colorectal surgery [14].

The multiple regression analysis showed that minimally invasive surgery positively impacted on postoperative recovery (p = 0.004). Despite a positive impact of minimally invasive technique on postoperative outcomes has been widely demonstrated in elective colorectal surgery [15, 16], the implementation of laparoscopy in emergency surgery is still a matter of debate [17, 18]. In our cohort about 40% of patients had a minimally invasive approach and laparoscopic surgery was successfully completed in two-thirds of them. A national UK study showed that laparoscopy was adopted in less than 20% of patients who underwent emergency surgery [17]. The present data should contribute to a wider use of minimally invasive approach in emergency surgery.

Intraoperative fluid management is a cornerstone of enhanced recovery protocols. In elective major noncardiac surgery a goal-directed fluid management strategy reduced postoperative complications [19, 20]. In the present study, a fluid overload negatively impacted on postoperative recovery delaying mobilization and oral feeding (p < 0.001). To prevent the risk of fluid overload, operative hemodynamic monitoring should be implemented to yield a proper goal-directed fluid therapy [21]. An abdominal drain was placed in 55.7% of patients operated for an intestinal obstruction, suggesting an over-indication in absence of a macroscopic peritoneal contamination. Similarly to the elective setting [22], the abdominal drain was negatively associated with postoperative recovery (p = 0.007), therefore, it should be placed in selected cases and removed as early as possible.

Preoperative hyperglycemia has been strongly correlated with morbidity and mortality in both elective [23,24,25,26,27] and critically ill surgical patients [28,29,30]. Hyperglycemia could also be considered a marker of severity of disease and organ dysfunction. Our data confirm the importance of a tight glycemic control in patients undergoing emergency surgery (p = 0.002). The repair of perforated peptic ulcer was associated with delayed postoperative recovery, too (p < 0.001). Despite randomized trials reported both feasibility and safety of an enhanced recovery protocol after peptic ulcer repair [4, 5], our results probably reflect some reluctance of surgeons to early feed these patients.

The present study has some limitations. Patients were recruited during the second pandemic wave with all the well-known restrictions and changes in hospital admissions and clinical practice. Pandemic affected hospitals’ organizational models making more difficult to have a dedicated equipe for emergency patients management. Moreover, this study is burdened by the intrinsic limits of the emergency setting not allowing a fixed and dedicated team with possible incomplete protocol application. On the other hand, strengths of the study are the large number of patients recruited in one-year period and the wide experience of the participating centers in enhanced recovery practice. There are several areas of possible improvement, such as preoperative hyperglycemia correction, operative fluids optimization, and implementation of minimally-invasive approach.

In conclusion, the present study supports the implementation of enhanced recovery protocols in patients undergoing emergency gastro-intestinal surgery. Multiple regression analysis showed that laparoscopy was associated with an earlier recovery, whereas preoperative hyperglycemia, fluid overload, and abdominal drain were associated with a delayed recovery.