Timing of Postoperative Drain Removal Following Parotidectomy - a Prospective Randomized Controlled Study

Overview

There is no standard of care of management following parotidectomies. Most practitioners however place a drain (usually a flat Davol closed suction drain/ Hemovac in the investigators institutions) in the wound bed at the end of the procedure, which is usually removed the day following the surgery, or when the drain output reaches a subjective number (depending on surgeon preference). The investigators hypothesis is that there is no difference in hematoma/bleeding or infection rate when post parotidectomy drains are removed in the recovery area on the day of the surgery prior to discharge, compared to drains that are kept in place for at least 1 day or until drainage is less than 50 cc/24 hrs. The investigators would therefore like to undergo a randomized control clinical trial to assess the rate of significant hematoma formation and infection, the need for readmission and length of admission secondary to these complications when comparing 2 groups: one where drains are removed prior to discharge from the recovery area and the other where the patients are discharged with the drain and seen the next post operative day in the office. The drain output will then be measured and the drain taken out.

Background: Parotidectomies involve the removal of all or a portion of the parotid gland off the facial nerve in the case of superficial parotidectomies, and total removal of the gland in the case of total parotidectomies. The most common indications for this procedure are for benign neoplasms such as pleomorphic adenomas, oncocytomas and Warthin tumors, or for malignant involvement of the gland such as metastatic cutaneous squamous cell carcinomas. Hematoma formation or bleeding following parotidectomy is a less emphasized early complication of this procedure, usually secondary to unrecognized incomplete hemostasis prior to incision closure. Hematoma formation rate post parotidectomy has been reported to vary between 3.8% (Upton et al.) and 6.1% (Laccourreye et al.). In case of bleeding or rapidly expanding hematomas, the patient is usually brought back to the operating room for surgical drainage of the hematoma and surgical control of any identified bleeding vessels (Myers et al). Small hematomas can usually be managed expectantly. Wound infection following parotidectomy usually also ranges around 3.8% (Upton et al.) to 5.4% (Henney et al). Hypothesis: There is no difference in hematoma/bleeding or infection rate when post parotidectomy drains are removed in the recovery area on the day of the surgery prior to discharge, compared to drains that are kept in place for at least 1 day or until drainage is less than 50 cc/24 hrs. Justification for study: Review of the literature has shown paucity of publication about this particular complication and post operative preventative management. The current postoperative management following parotidectomies is not evidence based, and is institution or surgeon dependent, ranging from primary closure without a drain, to placement of a drain that is removed either on post-operative day 1-14 (in hospital or in the community), and from no overnight stay to patient hospitalization until the drain is removed. Patel et al. 2006 have published a review of postoperative drainage following their parotidectomies, including primary closure without drain, with drain, with absorbable hemostatic agents and with fibrin sealants. As mentioned by these authors, suction drains do not prevent hematomas, but do help to remove drainage and close dead space. In his paper, the average drain output on post operative day 1 was 27.1 ml, but these authors do not address the optimal time for drain removal. The timing of removal of neck drains following head and neck surgery was addressed by Harris T. et al, who found that drains could be safely removed when the volume falls to 50 ml or less over a 24 hour period, with only 9% of their patients developing seromas. Moreover, a retrospective review of the timing of neck hematoma formation post carotid endarterectomy showed that 95% of their patients experiencing neck hematomas or neurological deficits occurred within 8 hours following surgery, indicating that same evening discharge was possible following this type of surgery (Sheehan M.K et al). In their prospective four armed randomized controlled trial investigating the use of subcutaneous catheter drain and antibiotic irrigation of abdominal incision closures, Farnell et al. have found no statistical difference in infection rate in wounds with lesser degree of contamination (Type II and III) between groups without drains and with drains for 4 days. Clean procedures (Type I wound) such as in parotidectomies should thus have similar infection rates with very temporary drains compared to no drains. The optimal time for catheter removal was however not addressed in this study. The primary investigator has noted during his 30 years of practice as a Head and Neck surgeon that significant hematoma formation or bleeding that would require surgical management following parotidectomy is usually a very early complication that is often recognized within the few hours following surgery due to the superficiality of the wound, when the patient is in the postoperative area. It is our belief that removal of the drain in the post operative area, just prior to the patient's discharge allows for the decreased risk for unrecognized bleeding while and decreasing risks of nosocomial infections and patient colonization with resistant bacterial strains, unnecessary prolonged discomfort to the patient in the form of the drain, as well as unnecessary use of hospital resources such as the need for overnight stay in the hospital. The average cost for one postsurgical hospital day for an out of province resident at Vancouver General Hospital costs about C$3000 per day. We also believe that wound infection rate would be equivalent if not less. Clinical equipoise exists as there is still controversy about the timing of the drain removal. Since no previous study has addressed the need or optimal timing of surgical drain removal following parotidectomies, we would also like to undertake a randomized controlled trial to address the hematoma and infection rate in post parotidectomy patients, depending on whether a drain is removed in the recovery area, or the next post operative day. This would follow a retrospective review study (H11-01433). Objectives: In our prospective randomized controlled trial, we would like to assess the rate of hematoma formation and infection, the need for readmission and length of admission secondary to these complications when comparing 2 groups: one where drains are removed prior to discharge from the recovery area and the other where the patients are discharged with the drain and seen the next post operative day in the office. The drain output will then be measured and the drain taken out. Research Methods: * A prospective randomized controlled trial: * Inclusion criteria: patients age ≥ 18 years old, competent, undertaking unilateral or bilateral parotidectomies at Vancouver General Hospital, Vancouver; St. Paul's Hospital, Vancouver; or Surrey Memorial Hospital, Surrey, Eagle Ridge Hospital, Port Moody. * Exclusion criteria: patients on anticoagulation therapy, ASA ≥ 3 where a patient would require post operative overnight admission secondary to his comorbidities, immunosuppression such as concurrent radiation or chemotherapy or receiving high dose corticosteroids, and concommitant infections requiring antibiotics, patients who are having additional head and neck procedures at the time of the parotidectomy, other than skin excision (eg. Modified neck dissection, mandibulectomy, external auditory canal resection) * Patients will be recruited and consent obtained (see below for details of consent). * The consented patients will be randomly assigned a unique subject number, not derived from personal identifiers, then randomly assigned to either of the 2 arm groups, through the use of opaque envelopes that will be opened after skin closure: 1. Drain placement and patient discharge with the drain. The patient will then be seen the next postoperative day by the surgeon to have the drain output measured and the drain taken out. In the very unlikely case that a drain output exceeds 50 cc/24 hrs, the patient will be asked to come back on the 3rd post operative day for drain removal, when the output will likely be even less than 50cc/24 hrs (Harris, T et al.). 2. Drain placement with removal prior to discharge on the same postoperative day from the recovery area, irrespective of volume drained (volume will be measured). * Permutated blocks of 4, unbeknownst to the surgeon will be used to allow for equal balance in sample size of the 2 groups. * The patients will be stratified on the basis of which hospital they had the surgery at. * The patients and surgeon will be blinded until the end of surgery. The outcome assessor cannot be blinded to the different arms under this setting. * If clinical judgment requires that a patient be admitted, or have the drain kept longer than the assigned time, the patient will be analyzed on an intention to treat basis. * Outcomes measured include: * Significant Hematoma /Bleeding rate (primary outcome) * Infection rate * Length of hospitalization * Need for surgical reexploration * Definitions: * Significant Hematoma: Blood collection/Bleeding at the wound site that was diagnosed by a doctor and that requires surgical management. * Infection: Record of positive pus culture if patient in hospital, record of abscess diagnosed by a physician through clinical exam, imaging or during surgical drainage, or spontaneous drainage of purulent material from wound incision with associated localized pain. Wound erythema that spontaneously resolved will not be included in this outcome. * Length of hospitalization: Number of nights patient stayed in hospital. * Surgical reexploration: Patient is brought back to the OR for hemostasis or where wound incision is opened and drained following hematoma or abscess formation. * The outcomes will be measured by the outcome assessor at 30 days post operatively, where outcomes will be determined by contacting the patients (by questionnaire, telephone interview), regarding presence or absence of significant hematoma, infection, number of nights spent in hospital and whether or not they had a surgical exploration, with the definitions of these outcomes given to them. These outcomes will be confirmed via the patient's chart review or with the surgeon. * Details of consent: * Source of the contact information: * From the primary health care provider's file (the primary investigator). * Who will collect the contact information: * The investigator not acting as health care provider will obtain the list of patient's name from the VGH ENT outpatient clinic after they are consented for parotidectomies. * Who will make the initial contact with the prospective subject and how and when will the initial contact be made: * Non-coercive initial contact with the prospective participants will be undertaken through numerous approaches: notices to invite prospective participants will be posted in the clinic rooms, waiting rooms as well as the the secretary's office in the outpatient VGH ENT Clinic. Interested prospective participants can either call the investigator, who will mail the information package to their address or the secretary (acting as study co-ordinator) who can hand them an information package. * The secretary (acting as the study co-ordinator) will also hand a package to prospective potential participants. The package will contain an introductory letter signed by the Primary Investigator describing the background behind the study, justifications and aim of the study as well as the methodology of the study and details on how to contact the investigator. In the letter, they will also be notified that the investigator (not involved in their primary care, along with her stated relationship to the VGH ENT clinic and the Primary Investigator) will be contacting them within the next 1-2 weeks to give more information about the study, answer any questions that they might have and see whether they would be interested in participating in the study. They would also be invited to call the Primary investigator/ health care provider if they have any questions to which the person making initial contact cannot respond. Additionally, it will be confirmed in writing that the patient's decision in participation will have no impact on the patient's level of care. Also included in the package will be a consent form for the study. * On handing the package, the secretary/co-ordinator will tell the prospective participant that their health care provider and his colleague are undertaking a study involving the surgery they will be undergoing and that they are invited to learn more about the study through the package. The co-ordinator will also let them know verbally that Primary Investigator's colleague, who also works in the clinic, will be contacting them within the next 1-2 weeks to discuss the project in more detail and see whether they would be interested in the study (this will also be mentioned in writing in the introductory letter). * If the investigator (not involved in the care of the patient) is in the clinic at the time that the prospective participant is in the clinic, she will introduce herself to the prospective participant and detail her relationship to the VGH ENT clinic and the Primary Investigator, and hand the package to the patient, inviting them to go through the details of the study and informing the patient that she will call them within the following 2 weeks to discuss the study in more detail and to see whether they would be interested in participating. * The relationship of study team members to prospective participants: * The prospective participants will be under the Primary Investigator's care; therefore, the PI will not make the initial contact as detailed above. * The other investigator involved in the study (PGY2 ENT resident) will not be involved in the care of the patient, nor is the secretary/acting as co-ordinator and therefore will not create a coercive contact with the prospective participants. * The consent form will be explained by the investigator not involved in the care of the patient (but also working at the VGH ENT clinic), via the phone, within 1-2 weeks following the prospective participants' receipt of the information package (that includes the consent form), obtained at the time of their clinic visit or mailed to them. * The consent will be obtained after the phone discussion about the study, when the patient will be asked to sign the consent form included in their information package during their clinic visit and mail the consent in a prepaid envelope. If the patients decide to think about whether they want to participate, they will be invited to do so and bring the consent form on the day of the surgery. The procedure will be undertaken over the phone since no participant recruitment can take place at the time of the patient visit to the clinic, and so as to minimize patient inconvenience in the form of mobilization (unless prospective participants prefer to do so) for a face to face discussion. * During the phone call, the patients will be given information concerning the background behind this study, and justifications as well as the objectives for undertaking this study. The patients will be advised that, were they to consent to participate in this study, they would not know which arm they would be assigned to until after the surgery. * The patients would be reassured that their decision in participating in the study would not impact in their level of care. * They will be informed that were they to be randomized in arm #1, they would be given an appointment time for the following day to meet one of the authors for drain output measurement and drain removal. In the very unlikely event that the drain output is \>50 cc/24 hrs, another appointment for the post operative day 3 will be arranged for removal and measurement of drain output. * They would also be informed that one of the authors of the study would contact them at the 30 day mark to ask about whether they had any bleeding/significant hematoma, infection and number of nights spent in hospital and whether or not they had a surgical exploration, with the definitions of these outcomes given to them. They would also be informed that the obtained information would be confirmed through their hospital and medical records. * Time dedicated to discussion of study and prior to consent: 15 minutes unless patients has additional questions; this will take place within 2 weeks of the receipt of the information package. * Maximum time needed to dedicate to the study if patient is randomized to the group where the drain is removed the same post operative day: 1 minute. * Maximum time needed to dedicate to the study if patient is randomized to group where drain is removed on the next post operative day: 15 minutes (excluding transportation time) * Maximum time required for questionnaire filling/over the phone questionnaire (excluding time required to post prepaid envelope): 5 minutes; this will be taking place 1 month following the surgery. * Potential benefits to the subject that could arise from his or her participation in the proposed research: Possible therapeutic benefits include more frequent follow- up visits (and therefore patient reassurance) in the control arm. * Provisions made to break the code of a double-blind study in an emergency situation, and indicate who has the code: There will be no need for provisions to unblind the patients or surgeon as both arms will have drains placed in. The surgeons and patients will both be unblinded at the end of surgery. * All paper form study data including consent forms will be held in a locked filling cabinet in the ENT Department at Vancouver General Hospital. Patients will be assigned a random and unique number in a patient log, which will be kept in an appropriately secured electronic form. Data collected are identified using the unique subject number that has not been derived from any personal identifiers; i.e. the data collected is already de-identified before being entered in the database. This file containing the data collection is also password protected. Provisions during the consenting process: No provision is possible during the consenting process as the consent will be discussed with the patient on the phone. Sample size \& Statistical Analysis: 400 patients will be recruited to achieve a power of 0.8, taking into account that an intention to treat analysis will be undertaken, where it is assumed that average rate of hematoma formation post parotidectomy in the late drain removal group is 10% and 2.5% in early drain removal group. * At the 100th patient, an interim analysis will be undertaken, and if there is any significant difference in hematoma or infection rate (p\<0.05), the study will be stopped. Additional patients who would already be enrolled but have not been randomized yet would be contacted by writing to be advised of the results of the study and of the end of the study. * The results will then be anonymously analyzed using a t-Test by a medical statistician. The results of this study will be published in a medical journal. The data will be kept for a minimum of five years following publication of the study results, following which hard copies will be shredded and computer databases will be deleted.