Surg Endosc (2004) 18: 1463–1468 DOI: 10.1007/s00464-003-9238-y
Springer Science+Business Media, Inc. 2004

‘‘Fast-track’’ multimodal rehabilitation program improves outcome after laparoscopic sigmoidectomy A controlled prospective evaluation W. Raue, O. Haase, T. Junghans, M. Scharfenberg, J. M. Mu¨ller, W. Schwenk Department of General, Visceral, Vascular, and Thoracic Surgery, Medical Faculty, Humboldt University, Charite´, Campus Mitte, Schumannstrasse 20/21, 10117 Berlin, Germany Received: 25 August 2003/Accepted: 17 March 2004/Online publication: 26 August 2004

Abstract Background: Laparoscopic colorectal resection improves patient outcome by reducing pain, postoperative pulmonary dysfunction, gastrointestinal paralysis, and fatigue. A multimodal rehabilitation program (‘‘fasttrack’’) with epidural analgesia, early oral feeding, and enforced mobilization may further improve the excellent results of laparoscopic colorectal resection, enabling early ambulation of these patients. Methods: Fifty two consecutive patients underwent laparoscopic sigmoidectomy with standardized regular perioperative treatment (standard) or multimodal rehabilitation program (‘‘fast-track’’). Outcome measures included pulmonary function, duration of postoperative ileus, pain perception, fatigue, morbidity, and mortality. Results: Twenty nine standard-care patients (19 men and 10 women) and 23 fast-track patients (15 men and eight women) were evaluated. Demographic and operative data were similar for the two groups. On the 1st postoperative day, pulmonary function was improved (p = 0.01) in fast-track patients. Oral feeding was achieved earlier (p < 0.01) and defecation occurred earlier (p < 0.01) in the fast-track group. Visual analogue scale scores for pain were similar for the two groups (p > 0.05), but fatigue was increased in the standard-care group on the 1st (p = 0.06) and 2nd (p < 0.05) postoperative days. Morbidity was not different for the two groups. Fast-track patients were discharged on day 4 (range, 3–6) and standard-care patients on day 7 (range, 4–14) (p < 0.001). Conclusion: Multimodal rehabiliation can improve further on the excellent results of laparoscopic sigmoidectomy and decrease the postoperative hospital stay.

Correspondence to: W. Schwenk

Key words: Laparoscopic sigmoidectomy — Perioperative treatment — Pulmonary function — Morbidity — Hospital stay

The laparoscopic approach improves the postoperative course of patients undergoing elective colectomy. In randomized controlled trials, laparoscopic colectomy has been shown to reduce postoperative pain, the analgetic requirement, fatigue, and the duration of gastrointestinal paralysis [9], as well as improve pulmonary function and quality of life [13, 14, 18]. Despite these advantages, general complications, such as pneumonia or adverse cardiovascular events, still occur in 10% of these patients, and the duration of the postoperative hospital stay in Germany is 10–12 days [10]. Bardram et al. have shown that a multimodal rehabilitation program that includes epidural analgesia, early oral feeding, and enforced mobilization improves the results of laparoscopic colorectal resection and may reduce the postoperative hospital stay to 2 days [3, 4]. Because patients benefiting from such a ‘‘fast-track’’ rehabilitation program have not yet been compared with patients receiving regular standard care in the same institution, we performed a controlled clinical evaluation.

Patients and methods The assessment of the patient was prospective and controlled. When the fast-track program was initiated in our department, it was decided in the beginning to include only patients treated by one surgeon (W.S.). Therefore, the assignment of patients to regular standard treatment or ‘fast-track’ therapy was based on their admission to different surgical wards in the same institution. All patients undergoing elective laparoscopic sigmoidectomy between 21 October 2001 and 9 May 2003 were included. There was no other selection of patients to either the standard or the ‘fast-track’ group. Exclusion criteria were primary

1464 Table 1. Protocol for anesthesia, surgery, and postoperative rehabilitation after elective laparoscopic sigmoidectomy with standard care vs multimodal rehabilitation (fast-track) Time

Standard-care protocol

Fast track protocol

Preoperative Intraoperative

Informed consent Thoracic combined EDA (LA/opioid) if recommended by anesthesiologist five-trocar laparoscopy; extract nasogastric tube after surgeon’s order Admit to SICU; i.v. fluids according to body weight (2,000–2,500 cc); continuous EDA (LA/opioid) or systemic opioid pca, metamizol 4 · 1 g i.v.; nil per mouth; sitting up in bed

Informed consent; plan to discharge on day 3 after surgery Valdecoxib with premedication; thoracic combined EDA (LA/opioid) recommended by surgeon to all patients five-trocar laparoscopy; extract nasogastric tube at extubation Admit to regular nursing floor via PACU, limit postop. i.v. fluids to 500 cc continuous EDA (LA/opiod), parecoxib; 40 mg i.v., avoid systemic opiods, magnesium oxide 3 · 300 mg per day until first bowel movement; tea (max. 1,500 cc), yogurt 2 · 150 g; short walk outside of room; mobilized to chair for 2 h Continuous EDA (LA/opiod), valdecoxib; avoid systemic opiods; regular hospital food, drink >1500 cc; mobilized out of bed >8 h; walk outside of room twice; extract intraabdominal drain and urinary catheter

Day of surgery

Postoperative day 1

Postoperative day 2

Postoperative day 3

Postoperative day 8

Admit to regular nursing floor, continuous EDA (LA/opiod) or systemic opiod pca, metamizol 4 · 1 g i.v.; clear liquids and soup; additional i.v. fluids as ordered mobilized out of bed recommended; extract intraabdominal drain and urinary catheter Continuous EDA (LA/opiod) or continue systemic opiod pca, metamizol 4 · 1 g i.v.; mashed food; additional i.v. fluids as ordered; mobilization out of bed recommended Extract continuous EDA (LA/opioid) or discontinue systemic opioid pca, metamizol 4 · 1 g; regular hospital food; additional i.v. fluids as ordered mobilization out of bed recommended but not enforced Patient discharged following surgeon’s orders; discuss result of histological examination; plan adjuvant therapy if needed before patient is discharged

Extract EDC and CVC in the morning; valdecoxib; avoid systemic opiods, regular hospital food, drink >1,500 cc; fully mobilized, plan discharge Oral valdecoxib; avoid systemic opioids; regular hospital food, drink >1,500 cc; fully mobilized out of bed; discharge following patient’s wishes Outpatient clinic; extract skin staples; discuss result of histological examination; plan adjuvant therapy if needed

EDA, epidural analgesia; LA, local anesthetics; SICU, surgical intensive care unit; PCA, patient-controlled analgesia; PACU, postanesthesia care unit; EDC, epidural catheter; CVC, central venous catheter

conventional resection or conversion to formal midline laparotomy, age <18 years, or pregnancy. All ‘fast-track’ patients were operated on by one surgeon (W.S.), whereas patients from the ‘standard’ group were treated by one of three other surgeons. All surgeons were experienced in laparoscopic and conventional colorectal surgery. Details of the perioperative care programs are given in Table 1. All patients underwent a standard bowel preparation (Prepacol, Guerbet, Germany; 2I polyethylenglycol solution, Klean-Prep, Norgine, Germany) and received ‘single-shot’ antibiotic prophylaxis (1.5 g Cefuroxim, Zinacef, Glaxo Wellcome, Germany) and 0.5 g metronidazol (Clont, Bayer, Germany) at induction of anesthesia. All patients had a central venous line, an arterial catheter, and a nasogastric tube inserted after induction of anesthesia. Pneumoperitoneum was established with a Veress needle, and laparoscopic sigmoidectomy was performed using a five-trocar technique, as described by Milsom and Bo¨hm [11]. Specimens were extracted using a waterseal bag through a mini-laparotomy in the left lower quadrant. Due to severe inflammatory adhesion, dissection of the lymphovascular pedicle and mobilization of the left colonic flexure were performed laparoscopically, while the sigmoid colon was dissected via a low transverse abdominal incision. These operations were classified as laparoscopic-assisted sigmoidectomies and are included in the present series. All anastomoses were accomplished in double-stapling technique. A 24 Charriere Robinson drain was placed in the lower abdomen and removed on the 1st postoperative day. Pulmonary function was measured on days 1–3 and day 8 by bedside spirometry (Renaissance Spirometer, Puritan Bennett Hoyer, Jolstein, Germany) with the patient lying in bed and the upper body elevated 45 [1]. Patients who had already been discharged were seen in the outpatient clinic on the 8th postoperative day. At the same time, pain at rest, pain while coughing, and fatigue were assessed using a 100-mm visual analogue scale (VAS) [5, 6]. Until the 2nd postoperative day in ‘fast-track’ patients, continuous administration of fixed combination of opioids and local anesthetic (sufentanil/ropivacaine) via a thoracic epidural catheter (EDC) was preferred. Additionally COX II inhibitors were given (parecoxib 40 mg per day i.v. or valdecoxib 20 mg twice a day postoperatively). Systemic

application of opioids was avoided. If an EDC was contraindicated, piritramid was administered on demand to complete the basic medication of COX II inhibitors and metamizol every 6 h. Patients in the standard treatment group also received an EDC whenever possible; alternatively, piritramid was given as patient-controlled analgesia. Pain treatment was completed with 1 g metamizol on demand. In all patients, a VAS score at rest of <35 was the objective.

Data analysis Continuous parameters are given as medians (5th–95th percentile). To avoid multiple testing of repeated measures, area-under-the-curve (AUC) values were calculated [2]. Differences between groups were compared using the Mann-Whitney U test [2]. Categorical data were analyzed using Fisher’s exact test. A p value of <0.05 was considered significant. Data in figures are given as median (joined by line), 25th– 75th percentile (blocks), and 5th–95th percentile (whiskers). Statistical analysis of all data was performed using SAS 8.0 for Windows 98 (Microsoft Corp., CA).

Result Patients Fifty two consecutive patients underwent laparoscopic sigmoidectomy at the Department of General, Visceral, Vascular, and Thoracic Surgery of the Charite in Berlin. According to their admission to one of the two general surgical wards by general practitioners, 29 patients were treated according to a standardized perioperative care program (‘standard care’) and 23 patients underwent a

1465

multimodal rehabilitation program (‘fast track’). Epidemiological data of patients, indications for resection, type of resection, and operative technique are shown in Table 2. There were no differences between the two groups except for a higher percentage of female patients in the standard-care group. Duration of surgery was 236 (±62) min in the standard-care group and 182 (±44) min in the ‘fast-track’ group (p > 0.05).

Table 2. Epidemiological data, type of resection, and access after elective laparoscopic sigmoidectomy with standard care vs multimodal rehabilitation (‘fast-track’) Standard care (n = 29)

Fast-track (n = 23)

p-value

65 (38–86) 25.5 (21.3–33.3) 4.0 (2.6–5.6)

63 (32–76) 26.8 (22.1–29.7) 3.3 (2.0–4.6)

0.76 0.48 0.04

10/19 2/19/8/0

15/8 1/11/11/0

0.05 0.35 0.76

10 (34) 19 (66)

7 (30) 16 (70)

Preoperative pulmonary function was comparable in both groups (Fig. 1). Postoperatively, forced vital capacity (FVC) was better in the fast-track group on the 1st postoperative day, but no further differences between both groups were detected. Normal gastrointestinal function was achieved early in both groups, but fasttrack patients returned to a regular hospital diet earlier than patients receiving standard care (Table 3).

Epidemiological data Age (yr) BMI (kg m)2) Preoperative. FVC (I) Sex (F/M) ASA class Indication for resection Malignant Benign Concomitant disease Arterial hypertony Cardiac Pulmonary Diabetes Renal

Pain and fatigue

BMI, body mass index; FVC, forced vital capacity; ASA, American Society of Anesthesiologists Continuous data are given as median (5th–95th percentile)

Pulmonary and gastrointestinal function

13 7 5 3 1

(44) (24) (17) (10) (3)

9 7 7 2 1

(39) (30) (30) (9) (4)

0.78 0.76 0.32 0.99 0.99

Postoperatively, no differences were observed in the VAS score for pain during rest or while coughing. However, patients undergoing the fast-track protocol did suffer less fatigue on days 1 and 2 after surgery (Fig. 2). ‘Fast-track’ patients felt ready for earlier discharge than standard-care patients (Fig. 3). Morbidity, mortality, and readmission General complications occurred in three patients (11%) in the standard-care group (two urinary tract infections, one temporary hemiplegia after incidental carotid artery puncture) and one patient (4%) undergoing fast-track rehabilitation (hematuria). Local complications were diagnosed in three standard-care patients (11%) (two subcutaneous wound infections, one intraabdominal abscess) and none of the fast-track patients. There were no lethal complications, so the mortality rate was nil. Within 30 days after surgery, one patient (4%) of the standard group (local wound infection) and two fasttrack patients (7%) were readmitted. Readmission of both fast-track patients occurred for minor reasons (one patient ‘‘not feeling well’’, without specific diagnosis; one gastroenteritis), and neither patient required specific therapy.

Discussion The postoperative short-term benefits of laparoscopic colorectal resection have been demonstrated in several randomized controlled trials. Minimally invasive colorectal resection decreases pain and analgetic requirement [14], improves postoperative pulmonary function [15], shortens postoperative ileus and hospital stay [9], and improves quality of life [18]. Unfortunately, only a few of the randomized controlled trials published so far

Fig. 1. Perioperative changes in forced vital capacity (FVC) after elective laparoscopic sigmoidectomy with standard care (traditional) or multimodal rehabilitation (‘fast-track’).

have used modern techniques for pain relief [17], and in most trials conservative protocols for postoperative care were followed [9, 15]. Bardram et al. have introduced a multimodal rehabilitation program for patients undergoing laparoscopic and conventional colonic surgery that was able to reduce general morbidity and decrease postoperative stay in these patients [3, 4]. The main principles of this ‘‘fasttrack’’ program are (a) perioperative epidural analgesia with opioid and local anaesthetics, (b) enforced mobi-

1466 Table 3. Oral feeding, duration of postoperative paralytic ileus, and reinsertion of nasogastric tube after elective laparoscopic sigmoidectomy with standard care vs multimodal rehabilitation (‘fast-track’) Standard care (n = 29)

Reinsertion of nasogastric tube Tea/yogurt (postop. d) Solid food (postop. d) i.v. Fluids (postop. d) 1st bowel movement (postop. d) Discharge (postop. d) Readmission within 30 d

Fast-track (n = 23)

n

(%)

n

(%)

6 1 2 2 3 7 2

(21) (0–1) (1–3) (0–5) (1–6) (4–14) (7)

3 0 1 1 2 4 2

(13) (0–0) (1–1) (0–2) (1–4) (3–6) (9)

p value

0.7 <0.01 <0.01 <0.01 <0.05 <0.01 0.8

Continuous data are given as median (5th–95th percentile)

Fig. 2. Perioperative changes in fatigue Visual Analogue Score for Fatigue (VASF) after elective laparoscopic sigmoidectomy with standard care (traditional) or multimodal rehabilitation (‘fast-track’).

Fig. 3. Cumulative percentage of patients willing to be discharged after elective laparoscopic sigmoidectomy with standard care (traditional) or multimodal rehabilitation (‘fast-track’).

lization of the patient, and (c) early oral feeding beginning on the day of surgery [8]. After introducing this clinical pathway to their department, Bardram et al.

were able to shorten the postoperative rehabilitation period and discharge patients on the 2nd postoperative day after laparoscopic colonic resection. Even in elderly patients, mobilization after elective laparoscopic resection was rapid, such that patients were out of bed for =8 h on the 2nd postoperative day [4]. At the same time, gastrointestinal function was almost unimpaired after ‘fast-track’ laparoscopic colonic resection; 87% had an oral intake of >2,000 ml/day and defecation occurred in 94% of all patients within 48 h after operation [4]. The readmission rate (5%) was low in this series, and both readmissions were due to local complications (ileus and anastomotic leakage). Similar results were reported by Senagore et al. [16], who finished 159 of 181 intended laparoscopic sigmoid resections and discharged these patients 2.5 ± 1.2 days after surgery. General complications were infrequent (3.7%) in this series, and early gastrointestinal failure occurred in only 4.4% of all patients. Thirteen of the 181 patients (7.7%) were readmitted within 30 days with surgical complications (two cases of anastomotic leakage, two of pelvic abscess eight, of prolonged ileus two, of small bowel obstruction requiring surgery, and one mesenteric venous thrombosis). Although the results of these studies are encouraging, controlled clinical trials comparing the postoperative outcome of patients undergoing laparoscopic colon resection with different perioperative care regimens are rare. In a small randomized controlled trial, we did find a reduction in the use of intraoperative opioids but no differences in vomiting, duration of ileus, or time of discharge when low thoracic epidural analgesia was used in patients undergoing laparoscopic sigmoidectomy [12]. However, this trial had several drawbacks, as follows: (a) The epidural catheters were introduced at a low (Th 9–12) level, (b) epidural analgesia was achieved with a local anesthetic only, and (c) other aspects of the perioperative treatment were not altered [7]. The side effects of the pure local anesthetic epidural analgesia in this trial made early mobilization impossible in three of 10 patients. The present study is the first comparative trial to investigate the outcome of patients undergoing laparoscopic sigmoidectomy with a regular standardized care program vs a ‘fast-track’ program. The data show that a multimodal recovery program results in better pulmonary function, no gastrointestinal paralysis, and a

1467

shorter postoperative hospital stay when compared to a standard perioperative therapeutic regimen. The study was a controlled, nonblinded trial. There were two reasons for this study design: (a) Randomisation and blinding of patients and medical personal within one department to different postoperative care programs is difficult; and (b) when the ‘fast-track’ program was introduced into our department, it was decided to limit it to patients treated by one surgeon (W.S.) only. Therefore, the decision as to whether patients would undergo standard or ‘fast-track’ therapy was not made at random. Instead, allocation to a treatment group occurred by admission of patients to one of two general surgical wards of our hospital by general practitioners. As a result of this allocation, more female patients underwent ‘fast-track’ treatment, which explains the lower preoperative pulmonary function in this group. Because the treatment allocation was not randomized, possible biases related to preselection before admission or to the operating surgeon could not be extinguished. In fact, there was no influence of the wards on admission policy. All surgeons were well trained in laparoscopic and conventional colorectal surgery. In our department, laparoscopic sigmoidectomy is a highly standardized procedure; both written and videotaped operating pathways for this procedure were already in use beforehand. Moreover, all surgeons had performed >50 laparoscopic sigmoidectomies before the beginning of this study; therefore, the learning curve did not contribute to the differences between the two groups. The fast-track results of our study are comparable with the data reported by Bardram et al. [3, 4] and Senagore et al. [16]. Patients tolerated early oral feeding very well, early discharge was possible for most of the patients, general morbidity was low, and the readmission rate was acceptable. In contrast earlier studies [3, 4, 16], the reasons for readmission in the present trial were always nonsurgical. When the ‘fast-track’ concept was introduced into our department, indications for the readmission of these patients were very liberal. In fact, all patients who complained of any problem in the postoperative course after discharge from the hospital were readmitted. Therefore, we expect the readmission rate to decrease with growing ‘fast-track’ experience in our department. Although the results from the standard-care group were not as good as those from the fast-track group in this study, they still compare favorably to results from the literature concerning laparoscopic colonic surgery in Germany. Marusch et al. and the Laparoscopic Colorectal Cancer Study Group analyzed the data from 1,311 patients undergoing laparoscopic colonic resection by more (>100 laparoscopic resections; n = 813) or less (<100 laparoscopic resections; n = 498) experienced surgeons [10]. Specific complications occurred in 14.3% (more experienced) vs 15.1% (less experienced) of the patients, and general complications were diagnosed in 12.1% vs 9.0%. With regard to the experience of the surgeon, patients were discharged after a median of 12 vs 11 days after surgery. In the standard-care group of our study, the corresponding numbers were 14% (surgical complications), 11% (general complications), and 9

days. Therefore, the differences between the two groups truly reflect the improvement achieved by the multimodal rehabiliation concept. In summary, we conclude that the introduction of a multimodal rehabilitation program decreases pulmonary dysfunction, obviates gastrointestinal paralysis, and shortens the postoperative hospital stay after elective laparoscopic sigmoidectomy. However, when a ‘fast-track’ program is introduced into the clinical routine, a considerable number of patients may be readmitted within 30 days of operation. The different aspects of ‘fast-track’ laparoscopic colonic surgery (i.e., pain treatment, mobilization, and surgical technique) still need to be investigated in detail. Finally, the impact of a multimodal rehabilitation program on postoperative recovery after laparoscopic and conventional colorectal resections has to be evaluated in greater detail. Randomized, controlled, doubleblinded trials with a high level of evidence should be performed before routine administration of ‘fast-track’ treatment can be recommended for every patient undergoing laparoscopic colorectal surgery.

Acknowledgments. W. Schwenk is a member of the German Advisory Board for Parecoxib/Valdecoxib of the Pfizer Company.

References 1. American Thoracic Society(1987) Standardization of spirometry—1987 update. Am Rev Respir Dis 136: 1285–1298 2. Altman DG (1991) Practical statistics for medical research; 1st edition. Chapman & Hall, London 3. Bardram L, Funch-Jensen P, Jensen P, Crawford ME, Kehlet H (1995) Recovery after laparoscopic colonic surgery with epidural analgesia, and early oral nutrition and mobilisation. Lancet 345: 763–764 4. Bardram L, Funch Jensen P, Kehlet H (2000) Rapid rehabilitation in elderly patients after laparoscopic colonic resection. Br J Surg 87: 1540–1545 5. Christensen T, Bendix T, Kehlet H (1982) Fatigue and cardiorespiratory function following abdominal surgery. Br J Surg 69: 417– 419 6. Christensen T, Kehlet H (1993) Postoperative fatigue. World J Surg 17: 220–225 7. Kehlet H (2000) Randomized controlled trial to examine the influence of thoracic epidural analgesia on postoperative ileus after laparoscopic sigmoid resection. Br J Surg 87: 379 8. Kehlet H, Wilmore DW (2000) Multimodal strategies to improve surgical outcome. Am J Surg 183: 630–641 9. Lacy AM, Garcia-Valdecasas JC, Pique´ JM (1995) Short-term outcome analysis of a randomized study comparing laparoscopic versus open colectomy for colon cancer. Surg Endosc 9: 1101–1105 10. Marusch F, Gastinger I, Schneider C, Scheidbach H, Konradt J, Bru HP, Kohler L, Barlehner E, Kockerling F (2001) Experience as a factor influencing the indications for laparoscopic colorectal surgery and the results. Surg Endosc 15: 116–120 11. Milsom JW, Bo¨hm B (1996) Laparoscopic colorectal surgery. Springer, New York 12. Neudecker J, Schwenk W, Junghans T, Pietsch S, Bohm B, Muller JM (1999) Randomized controlled trial to examine the influence of thoracic epidural analgesia on postoperative ileus after laparoscopic sigmoid resection. Br J Surg 86: 1292–1295 13. Schwenk W, Bo¨hm B, Mu¨ller JM (1998) Laparoskopische oder konventionelle kolorektale Resektionen—beeinflusst die Operationstechnik die Lebensqualita¨t. Zentralbl Chir 123: 483–490 14. Schwenk W, Bo¨hm B, Mu¨ller JM (1998) Postoperative pain and fatigue after laparoscopic or conventional colorectal resections: a prospective randomised trial. Surg Endosc12: 1131–1136

1468 15. Schwenk W, Bohm B, Witt C, Junghans T, Grundel K Muller JM (1999) Pulmonary function following laparoscopic or conventional colorectal resection: a randomized controlled evaluation. Arch Surg 134: 6–12 16. Senagore AJ, Duepree HJ, Delaney CP, Brady KM, Fazio VW (2003) Results of a standardized technique and postoperative care plan for laparoscopic sigmoid colectomy. Dis Colon Rectum 46: 503–509

17. Stage JG, Schulze S, Moller P, Overgaard H, Andersen M, Rebsdorf-Pedersen VB, Nielsen HJ (1997) Prospective randomized study of laparoscopic versus open colonic resection for adenocarcinoma. Br J Surg 84: 391–396 18. Weeks JC, Nelson H, Gelber S, Sargent D, Schroeder G (2002) Short-term quality-of-life outcomes following laparoscopic-assisted colectomy vs open colectomy for colon cancer: a randomized trial. JAMA 287: 321–328