Background

Major colorectal surgery, i.e., surgery that involves wide resection of the colon and anastomosis, generally involves a prolonged hospital stay - on average 12-14 days. A stay of one week is usually the minimum that can be expected [1]. This prolonged occupation of a hospital bed is not usually owed to problems of morbidity but to the conventional care protocol followed. For decades this protocol has hardly been modified: it therefore does not take into account the advances that have been made in the perioperative management of such patients.

Inadequate pain management, intestinal dysfunction and immobilisation have been recognised since at least 1997 as among the main factors delaying postoperative recovery in patients subjected to major surgery [2]. This led Kehlet et al [3] to propose a series of measures designed to improve recovery following major colorectal surgery in their well-known multimodal recovery program. Currently, evidence-based multidisciplinary action protocols of this kind, known initially as fast-track or better called enhanced recovery programs (ERAS), are not achieving the degree of implantation hoped for [3]. Some authors suggest this to be due to the organisational demands they make on surgeons, anaesthetists and nursing staff, and to social, cultural and economic realities [4].

The results that can be achieved with ERAS - reductions in postoperative morbidity, average length of hospital stay and the consumption of resources - are, however, significant, and the general implantation of an ERAS for patients who are to undergo colorectal surgery is recommendable [5–8]. In some of the clinical practice guides available at http://www.reducinglengthofstay.org/, these programmes are considered to represent best clinical practice according to current scientific evidence. Unfortunately, the results communicated regarding surgery in an ERAS context have nearly all come from individual institutions, although they include those of four randomised clinical trial [5, 9–11]. Recently the results of an international study (five hospitals in different countries, four with no prior experience in ERAS) reporting on patient follow-up, the degree of acceptance and the degree of compliance with the protocol of a common ERAS, have also become available [12]. The results of these studies [5, 9–12] suggest that just making a protocol available is insufficient for objectives to be achieved; changes also need to be made to organisational strategies and the medical professionals involved in pre, intra and especially postoperative care require support, perhaps via continuing education.

The present work analyses the short and medium term results returned by an ERAS for colorectal surgery based on previously published protocols, followed at 12 Spanish hospitals.

Methods

Participating centres

The twelve participating centres were chosen for their organisational abilities and their experience and interest in colorectal surgery and patient care. These centres, distributed around Spain, ranged from large university hospitals to medium and small area hospitals (the Clínico from Zaragoza, La Paz, Clínico San Carlos and Gregorio Marañón from Madrid, General from Valencia, Mútua Terrassa, Do Meixoeiro from Vigo, Hospital d'Igualada, La Mancha Centro from Alcázar de San Juan, Universitario de Elche, Son Llatzer from Palma de Mallorca and Fundación Calahorra hospitals) with the collaboration of Health Technology Assessment Unit from Agencia Laín Entralgo, Madrid.

There were two meetings with at least two professionals of each centre, a surgeon and an anaesthetist. During these meetings discussions were held with national and international experts who assisted the group in the implementation of the program. These professionals were the persons in charge to develop the program in their centre and to starting the implementation of the protocol in two months after the second work session. There were two annual reunions with the group to supervise and improve the compliance of the protocol.

Study design

This prospective study, which involved 300 patients, was performed between 2008 and 2009. The ERAS used was developed by the authors on the basis of published protocols [3, 5, 8–12]. The variables recorded included perioperative mortality, length of hospital stay and compliance with the protocol.

The study was presented to the Hospital Ethical Board and accepted as this is an observational non-randomised study based on the best available evidence.

The research was conducted conformed to the Helsinki Declaration and to local legislation. Patients gave informed consent to participate in the study. This study has been registered in the ISRCTN register with the number ISRCTN16397735, you can also access to the registration information by the URL: http://www.controlled-trials.com/ISRCTN16397735.

Inclusion and exclusion criteria

  1. A)

    Inclusion criteria. All patients had to be 18 years of age or over and to be programmed for surgery for colorectal cancer without the need for a stoma or any further surgical procedure.

  2. B)

    Exclusion criteria: The need for emergency surgery, an American Society of Anaesthesiologists (ASA) class of IV, the need for a colostomy or ileostomy, the inability to provide informed consent, diabetes, slow evacuation previously documented by a digestive medicine unit.

Enhanced recovery program protocol

The ERAS required that, during the preoperative period, patients be given advice, that there be no preparation of the colon and that patients receive four carbohydrate-rich drinks (4 × 200 ml) one day prior to surgery plus two further drinks (2 × 200 ml) on the morning of surgery. During surgery, goal directed fluids were administered using oesophageal Doppler monitoring, and hypothermia and drainages avoided. After surgery, nasogastric tubes were not used, early mobilisation was practised, and oral fluids administered early. Table 1 summarises the protocol characteristics of the ERAS followed.

Table 1 Protocol characteristics of the followed enhanced recovery program
Full size table

Although not explicitly required by the ERAS, the following variables were also recorded: opioid-free pain control, the use of prophylactic medication for postoperative vomiting and nausea, and the use of epidural anaesthesia.

Patients were discharged following the criteria established by the ERAS protocol. All patients were followed-up for at least three months.

An online database http://www.ftsurgery.com/ was prepared for the collection of data from the different centres.

Data analysis

Dichotomous variables were recorded as absolute frequencies (number of cases) and relative frequencies (percentages). It should be noted that this work reports frequencies and percentages with respect to the data made available; therefore they do not always refer to all patients. Continuous variables were recorded as means and standard deviations (SD) or median plus maximum and minimum values, depending on whether or not their distribution was normal (determined by the Kolmogorov-Smirnov test). All analyses were made using 18 version SPSS software.

Results

Patients, surgery and postoperative treatment

Of all patients who initially met criteria for inclusion, 16 were excluded, 9 patients were in intensive care at the discretion of the anaesthetist and 7 received an unscheduled ileostomy. A total of 300 patients were finally included.

Table 2 shows the characteristic of the patients included, the surgical techniques used and the surgical procedures followed.

Table 2 Patient characteristics and surgical techniques and procedures followed
Full size table

The median age of the patients was 69 years (35-88); 51.5% were women (152 patients). The patients were distributed by ASA class as follows: I 10.2% (30), II 50.2% (148) and III 39.7% (117).

Some 63.6% (171) of the patients underwent laparoscopic surgery and 36.4% (98) open surgery. Some 15.3% of the laparoscopically intervened patients required conversion to laparotomy (22).

Sigmoidectomy and right hemicolectomy made up the majority of procedures performed (62.5%). Some 5.2% of patients (15) were subjected to preoperative radiotherapy. Laparotomy was medial in 88.8% and transversal in 11.2% of patients so intervened.

The mean preoperative hospital stay was 0.82 days (SD: 1.93). The mean duration of surgery was 155.4 min. The median length of time spent in the recovery room was 240 min (60-1500).

Main features of the protocol and their compliance rates

Table 3 shows the compliance rates for the main features outlined in the ERAS protocol, as well as for the other features not explicitly included in the protocol.

Table 3 Compliance with perioperative treatment and postoperative recovery after colon resection within the context of the enhanced recovery program followed
Full size table

The overall compliance to protocol was approximately 65%, but varied widely in its different components.

Postoperative hospital stay and readmission rate

The median length of postoperative hospital stay was 6 days (3-89 days). Figure 1 shows the distribution of length of postoperative stay. The mean postoperative length of stay according to surgical procedure was: right hemicolectomy and sigmoidectomy 8 days, left hemicolectomy 6 days, and transverse resection and subtotal colectomy 5 days.

Figure 1
figure 1

Length of postoperative stay of patients who underwent colon surgery within the context of the present enhanced recovery program.

Full size image

Some 2.7% of the patients (8) were readmitted following discharge for medical or surgical reasons. Three patients re-presented with febrile syndrome, two with nosocomial pneumonia, and three with pulmonary thromboembolism, diarrhoea with hyponatremia or abdominal wall abscess.

Some 7% of the patients (21) required repeat surgery. The causes included dehiscence (12 patients), fistula or anastomatic leakage (three patients), evisceration (two patients), and abdominal pain, haemoperitoneum, ischemia and intestinal occlusion (one patient each). The median post-repeat-operative length of hospital stay was 6 days (1-20).

Postoperative morbidity/mortality

Table 4 summarises the complications encountered. The most common were surgical (23.7% of patients; 71), followed by septic 11% (33) and other medical complications 9.7% (29). The most common surgical complications were wound infection (12%; 36), paralytic ileus (10%; 30; median paralysis time 3 days [2–8]), anastomosis leakage (4.3%; 13) and perioperative haemorrhage (1.7%; 5). The most common medical complications were respiratory distress (2.7%; 8), respiratory failure (2.7%%; 8) and cardiac arrhythmia or ischemia (1.7%; 5). Finally, the most common septic complications were abdominal abscess (4.3%; 13), urine infection (3.7%; 11), respiratory infection (3.7%; 11) and peritonitis (2.3%; 7).

Table 4 Local and general morbidity following colon resection within the enhanced recovery program followed
Full size table

Three patients (1%), all of whom required repeat surgery, died. Two of these died during their hospital stay due to multi-organ failure and sepsis, and one from cancer following discharge.

Symptoms delaying discharge

Table 5 records the symptoms that delayed discharge. Some 31.3% (94) of all patients presented some such symptom, 22% (66) suffered one, 6.3% (19) suffered two, 2.7% (8) suffered three, and 0.3% (1) suffered four. The most common symptoms were vomiting (12%; 36), dyspnoea (6.7%; 20) and fever of unknown origin (4.7%; 14).

Table 5 Symptoms delaying discharge
Full size table

Discussion

Each of the steps outlined in the present ERAS is based on scientific evidence. For example, patient education is reported to be important in the response to surgery. Now-classic studies [3] have shown that informed patients require less analgesia in the postoperative period and indeed experience significantly less pain than uninformed patients. More recent work has shown that adequate preoperative information reduces patient anxiety before surgery and may also hasten postsurgical recovery [3, 13, 14].

A number of studies on programmed colon surgery have brought into doubt the need for preoperative mechanical cleansing of the intestine [3, 15]. The need for strict preoperative fasting has also been recently questioned. Most clinical practice guides suggest a period of absolute fasting of between two and six hours, but recent studies indicate that taking a carbohydrate-rich drink before surgery may reduce the endocrine catabolic response and improve insulin resistance [3, 16], improving surgical results and hastening recovery. The present ERAS included the administration of carbohydrate-rich drinks (4 × 200 ml) one day prior to surgery plus two further such drinks (2 × 200 ml) on the morning of surgery. The protocol also included the administration of goal directed fluids made possible by the standard use of oesophageal Doppler monitoring [3, 17–21], temperature control to avoid hypothermia [3], and the non-routine use of a nasogastric tube; meta-analyses of several trials suggest the latter may reduce pulmonary complications [3, 22]. A further measure was the avoidance of routinely using drainage; several randomised trials have suggested that drainage is of no benefit [3, 23, 24]. Drainage can be avoided in most patients or limited to a short period, facilitating early mobilisation [3], a measure also called for by the ERAS followed. Finally, although taking food orally is commonly limited in the postoperative period, a number of studies have shown that it is safe even after colon surgery involving anastomosis [3, 25, 26]; it was therefore included in the present ERAS.

Additional variables not explicitly included in the ERAS were also measured: use of prophylactic medication for postoperative vomiting and nausea, use of epidural anaesthesia [3], and opioid-free pain control, it has been reported that opioid-free or opioid-reduced analgesia may hasten recovery [3].

When protocols such as the present are implanted, the goal is that there should be full compliance with all measures outlined. However, full compliance is commonly very difficult to achieve [12]. In the present work the overall compliance rate was 65%, but varied widely in its different components. Patients received information in nearly all cases, while compliance with the provision of early postoperative fluids seemed particularly difficult.

The items of the protocol with less compliance were early oral fluid administration, goal-directed fluid therapy and early mobilization. The reason why these items obtained different compliance with the protocol could be the taste of oral fluid, rejection by patients, unavailability of devices and temporary employment of some healthcare providers involved in the ERAS. Probably the implementation could improve involving and training all the professionals who assist the patients included in the protocol and identifying these patients with signboards on bedside.

Missing values were most important for the variable surgical approach with 6.2% of them, being lower in the other variables.

The rates of complications and mortality recorded were similar to those reported by other authors in randomized controlled trials [5, 8–11]. In the present work the most common complication was wound infection. These programmes do not, therefore, appear to place the patient at any extra risk. Our results are similar to previous multicenter studies [27, 28] in terms of surgical complications, mortality and readmission rate. The surgical complication rate was 24% compared to the 14.1% reported by Schwenk et al [27] and 20% reported by Braumann et al [28]. The mortality rate was 1% compared to the 0.8% reported by Schwenk et al [27] and 0,4% reported by Braumann et al [28]. The readmission rate was 2.7% compared to the 3.9% reported by Schwenk et al [27] and 4% reported by Braumann et al [28].

Finally, the use of the present ERAS was associated with a preoperative hospital stay of fewer than 24 h and an overall mean stay of 6 days. In other recently published Spanish multicenter study, including data from 50 hospitals, the mean postoperative stay after colorectal resection was 12.36 days [29].

Conclusion

The present ERAS posed no risk to patients in terms of morbidity, mortality and shortened their hospital stay. The present results show that these programmes can be of benefit to patients and, by reducing hospital costs, may benefit society as a whole.