Reduction of Postoperative Complication Rate with the Use of Early Oral Feeding in Gynecologic Oncologic Patients Undergoing a Major Surgery: A Randomized Controlled Trial


of 10

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Background A randomized controlled trial was performed to assess the outcome of early oral postoperative feeding (EOF) compared with traditional oral feeding (TOF) in gynecologic oncology patients undergoing a complex laparotomy, including upper
  ORIGINAL ARTICLE – GYNECOLOGIC ONCOLOGY Reduction of Postoperative Complication Rate with the Useof Early Oral Feeding in Gynecologic Oncologic PatientsUndergoing a Major Surgery: A Randomized Controlled Trial Lucas Minig, MD 1,6 , Roberto Biffi, MD 2 , Vanna Zanagnolo, MD 1 , Anna Attanasio, MD 3 , Carmen Beltrami, RN 4 ,Luca Bocciolone, MD 1 , Edoardo Botteri, MS 5 , Nicoletta Colombo, MD 1 , Simona Iodice, MS 5 , Fabio Landoni, MD 1 ,Michele Peiretti, MD 1 , Giovanni Roviglione, MD 1 , and Angelo Maggioni, MD 1 1 Gynecology Department, European Institute of Oncology, Milan, Italy;  2 Abdomino-Pelvic Surgery Department, EuropeanInstitute of Oncology, Milan, Italy;  3 Anesthesiology Department, European Institute of Oncology, Milan, Italy; 4 Gynecologic Nurses Department, European Institute of Oncology, Milan, Italy;  5 Division of Epidemiology andBiostatistics, European Institute of Oncology, Milan, Italy;  6 Division of Cancer Treatment and Diagnosis, National CancerInstitute, National Institutes of Health, Bethesda, MD ABSTRACTBackground.  A randomized controlled trial was per-formed to assess the outcome of early oral postoperativefeeding (EOF) compared with traditional oral feeding(TOF) in gynecologic oncology patients undergoing acomplex laparotomy, including upper abdominal surgery. Methods.  Patients aged 18–75 years, undergoing anelective laparotomy and with a preoperative suspicion of gynecologic malignancy, were eligible. Exclusion criteriaincluded infectious conditions, intestinal obstruction,severe malnutrition, American Society of Anesthesiologistsscore C 4, intestinal resection, and postoperative stay in theintensive care unit lasting [ 24 h. Patients allocated to EOFreceived liquid diet in the first postoperative day and thenregular diet. Patients received traditional feeding schemeuntil resolution of postoperative ileus to start liquid diet.The primary end-point of the trial was length of hospitalstay. Results.  Between January 1, 2007, and November 17,2007, a total of 143 patients were randomized to receiveeither EOF or TOF. Hospital stay for patients who receivedEOF ( n  =  71) was 4.7 vs. 5.8 days for the TOF group( n  =  72) ( P  =  0.006). The mean level of postoperativesatisfaction was significantly higher in the EOF group (82.8vs. 71.7 mm,  P  B  0.001). Patients who received the TOFscheme had significantly higher overall postoperativecomplications (39 vs. 17% in EOF group,  P  =  0.003) andinfective complications (14% in TOF group vs. 3% in EOFgroup,  P  =  0.017). Variables such as nausea and vomiting,analgesic and antiemetic requirement as well as level of pain and quality of life were not different between groups. Conclusions.  On the basis of these findings, the policy of EOF should be used after a complex gynecologic oncologiclaparotomy.Currently, surgical management constitutes the firsttherapeutic step of most gynecological cancers. 1 Tradi-tionally, oral feeding has been delayed after laparotomyuntil the ileus is clinically resolved. 2 Thereafter, a liquiddiet is administrated with a gradual progression to solidfood diet. Patients undergoing laparotomy often areassigned to at least 2–3 days of partial or total starvation.This policy is generally followed in the hope of avoidingserious complications, such as anastomotic leakage, aspi-ration, and wound dehiscence. 3 Early oral feeding (EOF) consists in the resumption of solid meals before postoperative resumption of intestinalactivity. Although the safety of this new approach in post-operative dietary management is supported by severalstudies of abdominal and pelvic surgeries, a survey of European surgeons shows that only 5–50% have imple-mented an EOF regimen. 4–9 The EOF in previousgynecologic studies has been evaluated in patients withdisease limited to the pelvis and in those who require pelvic   Society of Surgical Oncology 2009First Received: 2 April 2009;Published Online: 4 September 2009L. Minig, MDe-mail: lucasminig@yahoo.comAnn Surg Oncol (2009) 16:3101–3110DOI 10.1245/s10434-009-0681-4  and aortic lymphadenectomy due to oncologic disease. 10–12 We evaluated EOF in women who underwent complexgynecologic oncologic operations, including upper abdom-inalsurgery.Lengthofhospitalizationwasprimaryoutcome. PATIENTS AND METHODS Patients Patients aged 18–75 years, who were scheduled toundergo laparotomy at the European Institute of Oncology(IEO), Milan, Italy, on the basis of a preoperative diagnosisof gynecologic malignancy were eligible for enrollmentonto this trial. All patients were willing and able to givewritten informed consent, which was previously approvedby the IEO ethics committee (protocol S328/506) inaccordance with the ethical standards of the HelsinkiDeclaration of 1975, revised in 1983. The exclusion criteriawere divided into three categories: (1) preoperative,including infections, intestinal obstruction, severe malnu-trition, as defined by weight loss of  [ 10% of baseline in thepast 3 months, and American Society of Anesthesiologistsscore  C 4; (2) intraoperative, including total or anteriorpelvic exenteration and intestinal bowel resection; and (3)postoperative, including admission to the intensive careunit for [ 24 h, and final histopathologic diagnosis reveal-ing benign or nongynecologic disease. Perioperative Care In accordance with published guidelines, all patientswere evaluated for nutritional status by means of a sub- jective global assessment performed with a standardizedquestionnaire. 13,14 This included the patient’s history(weight loss, changes in dietary intake, gastrointestinalsymptoms, and functional capacity), physical examination(muscle, subcutaneous fat, sacral and ankle edema, ascites),and the clinician’s overall judgment of the patient’s status(normal, moderately, or severely malnourished). On thebasis of these data, patients were classified as well nour-ished, moderately nourished, or severely malnourished.In agreement with institutional protocol, all patientsunderwent bowel preparation, four doses of monosodicphosphate 41.732 g, bisodic phosphate 9.531 g (Phospho-Lax), and prophylactic antibiotics 30 minutes before sur-gery, ceftizoxime 2 g administered intravenously (IV). Anasogastric tube was placed in all patients until surgery wascompleted and was then removed. It was replaced aftersurgery only in patients with paralytic ileus. A thoracicepidural catheter (16 gauge) was placed at the T7–T10level (dose test, lidocaine 2%, 2 ml). All procedures wereperformed by staff surgeons, assisted by another attendingsurgeon, fellow, or resident, and always with the partici-pation of an abdominal oncology surgeon. Once surgerywas completed, and after verifying that no intraoperativeexclusion criteria were met, patients were randomized bymeans of the Web-based Tenalea randomization system( patients underwent general anesthesia. In addition, itwas the intention to place an epidural catheter before sur-gery in all patients who consented to the procedure andwho had no contraindications. Postoperative analgesia viaepidural catheter was used routinely for 3 days, unlessthere was an abnormality in coagulation studies, difficultyin catheter placement, or the patient declined. The baselineregimen was ropivacaine (2 mg/ml), and fentanyl (6–12 ml/h), via an epidural catheter until day 3. If the patientscomplained of pain, additional analgesics of ketorolac(30 mg IV, maximum three doses a day), and morphine (2–4 mg IV), were subsequently prescribed. On day 3, thecatheter was removed and ketorolac (30 mg IV, three timesa day) was infused. In patients without an epidural catheter,analgesia was provided as intravascular continuousadministration of 90 mg of ketorolac and 200 mg of tramadol over 24 h for the first postoperative day. The painwas controlled on the second and third day with 90 mg of ketorolac and 100 mg of tramadol. Postoperative emesiswas treated with dolasetron (12.5 mg/day IV). If nauseaand emesis persisted, dexamethasone (8 mg IV) andtransdermal scopolamine were subsequently administered.  Data Collection Nursing staff recorded analgesic/antiemetic drugrequirements and intensity of pain. Frequency of postop-erative nausea and vomiting was also recorded. Patientswere asked to report the first recognition of intestinalactivity, first gas, and stool passage during the hospitalstay. Further clinical data regarding postoperative compli-cations, length of hospital stay, level of satisfaction, andquality of life (QoL) were also gathered. Groups Patients were randomly assigned to receive either earlyoral (EOF) or traditional oral feeding (TOF). Patients in theEOF group were offered liquids, mineral water (no gas), tea,chamomile infusion, or apple juice during the first 24postoperative hours (day 0). If well tolerated, as evidencedby an absence of nausea and vomiting, they were then fed aregular diet of boiled or grilled beef, chicken, or fish startingon day 1 and for the entire duration of the hospital stay.Patients randomized to TOF scheme received nothing bymouth while they awaited the resolution of postoperativeileus. After resumption of normal bowel function, as 3102 L. Minig et al.  expressed by the presence of bowel sound and the passageof flatus, and if patients did not have nausea and emesis,they were switched to oral liquid diet for 24 h. If this reg-imen was well tolerated, they were then placed on asemisolid diet of rice or pasta, vegetable soup, boiledpotatoes, or bread for 1 day. Regular diet was finally pre-scribed, provided the semisolid diet had been well tolerated. End Points The primary end point in this study was the length of hospital stay. Other outcomes were recovery of intestinalactivity, assessed by time to first recognition of resumptionof bowel function and passage of gas and stool; intensity of abdominal pain; presence of nausea, and emesis; postop-erative complications; patients’ satisfaction level and QoL;and analgesic and antiemetic drug requirements. Patientswere discharged by an attending physician when they wereable to tolerate a regular diet for at least 24 h in the pres-ence of resumption of bowel function, and when they hadnormal clinical parameters and physical examination. QoL, Pain, and Satisfaction Assessment  Pain intensity referredbypatientswas assessedbya visualanalog scale (VAS), measured in millimeters, where 0 indi-cated the least abdominal pain and 100 indicated the worstabdominal pain. 15 Patients were questioned for pain intensityevery 8 h at 8:00 AM, 4:00 PM, and midnight daily aftercompletion of surgery until hospital discharge. QoL wasassessed by two separate tools, which were validated QoLquestionnaire (QLQ) instruments designed by the EuropeanOrganizationforResearchandTreatmentofCancer(EORTC)to measure overall health status in patients with cancer. Spe-cifically, we opted for the EORTC QLQ-C30 and EORTCQLQ-OV28 questionnaires. They were both administeredbefore surgery and 30 days after hospital discharge.EORTC QLQ-C30 is a self-administered questionnairethat can be completed in a matter of minutes and has beenproven to be a valid, discriminating, reliable, and internallyconsistent tool. 16 EORTC QLQ-OV28 is a tool developedby the EORTC to assess QoL in patients with ovariancancer. 17,18 This questionnaire specifically addressesabdominal complaints, menopausal symptoms, attitudetoward disease and treatment, body image, and sexuality.We used the EORTC QLQ-OV28 ovarian cancer ques-tionnaire as a supplement to the EORTC QLQ-C30because it contains a number of items relevant for patientsundergoing gynecologic oncologic surgery.Patient satisfaction was measured by VAS before dis-charge from hospital. 19 A 10-cm-long scale was used, withadjectival descriptions at the ends and intermediate posi-tions. Participants were provided with a questionnairecontaining the VAS, together with a statement explainingwhat the scale was intended to measure. This framingstatement was introduced after initial testing of the scaleyielded a high variability. Postoperative Complications The diagnostic criteria for each postoperative compli-cation were established. Wound infection was defined asany redness and tenderness of surgical wound with dis-charge of pus and positive culture for any bacteria;abdominal abscess as deep collection of pus and positiveculture for any bacteria; pneumonia as abnormal chestradiograph findings with fever (temperature [ 38  C) andwhite blood count [ 12,000 cells/  l l and positive sputum orbronchoalveolar lavage. Urinary tract infection was definedas [ 10 7 microorganisms/ml of urine; bacteriemia as twoconsecutive positive blood cultures without shock; andwound dehiscence as any dehiscence of the abdominalfascia [ 3 cm. Ileus was defined as presence of nausea,vomiting, abdominal distension, lack of bowel sounds, air/ fluids levels on abdominal film, delayed passage of flatusrequiring nasogastric tube placement; marked postopera-tive bleeding as if requiring blood transfusion (  C 2 units);anastomotic leak as any clinical or radiological evidence of intestinal dehiscence; respiratory failure as presence of dyspnea and respiratory rate [ 35 breaths/min or PaO 2  of  \ 70 mmHg; cardiovascular instability as unstable bloodpressure requiring use of extra fluids and/or pressors; renaldysfunction as increased serum urea and/or creatinineconcentration (50% above baseline) and thromboemboliccomplications were defined as stroke, deep vein thrombo-sis, and pulmonary embolism. Follow-Up Afterhospitaldischarge,weeklytelephonefollow-upsfor30 days recorded all occurrences of readmissions to otherhospitals and/or long-term postoperative complications. Statistical Analysis: Trial Design and Sample Size The primary end point of the trial was the length of hospital stay. The target population was composed of patients undergoing a gynecologic oncologic laparotomywithout intestinal resection. The sample size was calcu-lated from data on the target population in our institutionbefore starting the study. These patients had a medianhospitalization time of 5 days. We expected a reduction of 20% in the length of hospital stay for patients receivingEOF, with a standard deviation of 1.8, a power of 90% andsignificance level (alpha) of 0.05. Thus, the final numberrequired was 70 patients per arm. Early Feeding After Gynecologic Laparotomy 3103   Data Analysis The Mann–Whitney  U  -test and Kruskal–Wallis test wereadopted to compare means and medians, respectively. The v 2 or Fisher’s exact test were used to compare proportions.Log transformations were used when necessary to achievenormality. A univariate linear regression model was used tostudy the effect of potentially explanatory variables onhospital stay. In the multivariate analysis, the effect of thenutritional group on hospital stay was adjusted for variablesthat were statistically in the univariate analysis, and thedifference between the adjusted means (or least squaremeans) of the two nutritional groups was reported. Theanalyses were performed with SAS software, version 8.2(SAS Institute, Cary, NC). All tests were two-sided. RESULTS Characteristics of the Patients Between January 1, 2007, and November 15, 2007, atotal of 281 patients were screened. Forty-two patients metpreoperative exclusion criteria, and 10 refused to partici-pate. Of the 229 eligible patients, 62 were excluded as aresult of intestinal bowel resection. One hundred sixty-fivemet the criteria for randomization. Twenty-four patientswere subsequently excluded as a result of benign gyneco-logic pathology, nongynecologic pathology, and admissionto the intensive care unit for [ 24 h. A total of 143 patientswere finally included in the following analysis (Fig. 1). Nodifferences between the two feeding groups were observedin patient characteristics. However, it is interesting to notethat a pelvic lymphadenectomy was performed in [ 70% of patients, and an aortic lymphadenectomy was performed inalmost 50% of women. In addition, 19 patients whoreceived upper abdominal surgery were studied, and 60%of women had ovarian cancer. This represents the largestsample of ovarian cancer patients studied in a randomizedcontrolled trial of EOF to date (Table 1).  Hospital Stay Table 2 summarizes length of hospital stay. Patientswho received EOF showed a significantly reduced length of hospitalization time compared with patients who receivedTOF (4.7 vs. 5.8 days, respectively;  P  =  0.006). In themultivariate analysis, after adjusting for other factors thatalso modified the length of hospital stay, such as upper Assessed for eligibilityn = 281Randomizedn = 165Allocated to intervention (n = 84)Received intervention (n = 84) Analyzed (n = 72) Excluded from analysis 12 Benign path 5 Non-gynecologic malignancy 4 Admission to ICU ≥  24 hrs 3Lost in follow-up (n = 0)Allocated to intervention (n = 83)Received intervention (n = 83) Analyzed (n = 71) Excluded from analysis 12 Benign path 5 Non-gynecologic malignancy 2 Admission to ICU ≥  24 hrs 5Lost in follow-up (n = 0) Excluded Did not meet inclusion criteria 106 * Planned pelvic exenteration 16 ASA score ≥  4 5 Severe malnutrition 2 Preoperative infection 3 Preoperative intestinal obstruction 7 Intestinal resection 62Refused participation 10*10 patients did not meet multiple inclusion criteria Enrolment Analysis Allocation Follow-up FIG. 1  Flow diagram for thetrial.  ASA  American Society of Anesthesiologists score.  ICU  intensive care unit3104 L. Minig et al.  abdominal surgery, higher International Federation of Gynecology and Obstetrics stage, intraoperative and post-operative complications, late time to passage of flatus, andlong surgical time; the difference between the two feedinggroups remained significant ( P  =  0.019). Other variables,like age, body mass index, estimated blood loss, histolog-ical type of malignancy (ovarian versus nonovarian),nausea and emesis, bowel sound, time to first defecation,and postoperative epidural anaesthesia, did not statisticallysignificantly affect the effect of the nutritional group onlength of hospital stay. EOF Tolerance and Overall Postoperative Satisfaction Eighty-nine percent of the patients in the EOF groupwere able to resume solid oral intake on the first postop-erative day, with no statistically significant difference inthe incidence of nausea and vomiting compared with TOFgroup (Table 3). Fifty-eight percent of the patients in theTOF branch expressed their desire to resume oral feedingearlier. Table 3 also shows that the mean level of overallpostoperative satisfaction was significantly higher inpatients who received EOF,  P  = \ .001. TABLE 1  Patient characteristics and baseline dataCharacteristic TOF ( n  =  72) EOF ( n  =  71)  P  value a Age (years)Mean age (range) 57 (23–76) 54 (20–75) 0.195Body mass index (kg/m 2 )Mean (range) 25 (17–42) 25 (18–39) 0.922Previous surgery 38 (53%) 37 (52%) 0.937Preoperative chemotherapy (last 6 months) 19 (26%) 15 (21%) 0.460Surgical proceduresHysterectomy (any type) 61 (85%) 54 (76%) 0.192Omentectomy 42 (58%) 35 (49%) 0.278Pelvic lymphadenectomy 52 (72%) 50 (70%) 0.843Aortic lymphadenectomy 33 (46%) 30 (43%) 0.710Appendectomy 22 (31%) 16 (23%) 0.278Upper abdomen surgery 7 (10%) 12 (17%) 0.530Diaphragm striping 4 (6%) 6 (9%) 0.532Splenectomy 1 (1%) 3 (4%) 0.366Partial liver resection 2 (3%) 1 (1%) 1.0Cholecystectomy 0 (0%) 2 (3%) 0.245FIGO stage b Stage I–II 33 (50%) 36 (58%) 0.360Stage III–IV 33 (50%) 26 (42%)Epidural anaesthesia 48 (66%) 47 (66%) 0.953Surgical time (min) (mean  ±  SD) 176 ( ±  46) 174 ( ±  58) 0.833Estimated blood loss (ml) (mean  ±  SD) 256 ( ±  259) 296 ( ±  382) 0.463Presence of ascites 14 (19%) 20 (28%) 0.220Intraoperative complications 3 (4%) 9 (13%) 0.067Primary siteOvary 41 (57%) 42 (59%) 0.789 c Endometrial 17 (24%) 9 (13%)Cervix 11 (15%) 16 (23%)Other 3 (4%) 4 (5%) EOF   early oral feeding,  TOF   traditional oral feeding,  FIGO  International Federation of Gynecology and Obstetrics,  SD  standard deviation. Dataare presented as  n  (%) unless otherwise indicated a P  value from  t  -test,  v 2 test, or Fisher’s exact test as appropriate b Two patients in EOF group were not staged; 13 patients reported relapse (7 in EOF and 6 in TOF groups) c Ovarian cancer versus other primary siteEarly Feeding After Gynecologic Laparotomy 3105
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks