No funding.

Background: Ketorolac tromethamine is a nonsteroidal anti-inflammatory drug that has been shown to reduce breakthrough narcotic consumption following orthopaedic, neurosurgical, and obstetrical procedures. The morphine-sparing effect of ketorolac may reduce respiratory and central nervous system depression post operatively, while still providing adequate pain control for patients. Patients undergoing abdominally based microsurgical breast reconstruction may particularly benefit from ketorolac postoperatively, and ultimately be safely discharged from hospital sooner than their counterparts receiving standard of care.

Research Question: In adult women post-mastectomy, undergoing abdominally based microsurgical breast reconstruction, does a postoperative regimen of intravenous ketorolac tromethamine, in addition to standard of care, reduce the length of postoperative hospital stay compared with an intravenous sham saline regimen?

Study Design: A single center, explanatory, placebo-controlled, 1:1 allocation, 2-arm, parallel group, superiority, randomized, and double blinded, controlled trial.

Population: The study population includes all females >18 years old post mastectomy consenting to abdominally based microsurgical breast reconstruction.

Intervention and Comparator: The intervention will be ketorolac 30 mg IV every 6 hours postoperatively for 72 hours. The comparator will be a sham intravenous administration of normal saline.

Outcomes: The primary outcome is hospital length of stay postoperatively. Secondary outcomes include visual analog scale for pain, breakthrough narcotic consumption, surgical drain outputs, hematoma, and other complications.

Sample Size: Available data provided estimates for average length of hospital stay and standard deviation. A minimally clinically significant difference of 1 day was decided on due to expert opinion. Based on a power of 80% and alpha of 0.05, and inflated to account for attrition and efficiency losses, a total of 50 patients (25 per group) will be required for this study.

Breast reconstruction (BR) is commonly requested following mastectomy, and recent measures have been taken to ensure patient access is universal. Transferring autologous tissue from the abdomen using microvascular surgical techniques provides a reliable reconstruction method with high levels of patient satisfaction, and is routinely performed. However, the surgery is relatively complex and lengthy; operative times can exceed six hours and hospital stay ranges from 5-9 days. During each patient's postoperative course she must meet certain milestones before discharge. It has been shown that beyond the necessary monitoring of the viability of the free flap, pain and a patient's ability to mobilize are important modifiable barriers to discharge.

Pain control following autologous BR is multimodal, but predominantly relies on opioids. Narcotics provide effective analgesia, but have a well-known side effect profile that includes drowsiness, nausea, vomiting, sedation, central nervous system and respiratory depression. Each of these potential adverse effects limits patients' mobility after surgery. Respiratory depression is particularly disadvantageous, as these patients have undergone major abdominal musculature dissection that also disrupts their mechanics of breathing 10 . This may have the counterproductive effect of limiting patient mobilization, and ultimately delay discharge. In a health care system where resources are at a premium and costs are climbing, any inexpensive intervention that allows for an earlier and safe discharge is worthy of consideration.

No standard approach to postoperative pain control in free flap BR exists. Most protocols are institution dependent, and variably incorporate narcotics, nonsteroidal anti-inflammatory drugs (NSAIDs), acetaminophen and other agents. In addition, these medications are given via oral, intramuscular, or intravenous routes. Recent studies have investigated the use of transversus abdominis plane blocks as an analgesic adjunct, and patients are often provided with patient-controlled analgesia (PCA) pumps in the early postoperative period.

Ketorolac tromethamine is an NSAID, which is approved by Health Canada for short-term pain control (Toradol, Hoffmann-La Roche Limited, 7070 Mississauga Road Mississauga, Ontario). Ketorolac provides better pain control than the most commonly used NSAIDs, and its potency is comparable to that of morphine. However, it does not have the same listed side effects of drowsiness and respiratory depression as do narcotics. Surgeons often cite concern regarding postoperative bleeding when using NSAIDs, however recent reviews have convincingly demonstrated the safety for use of NSAIDs in plastic surgery. A previous observational study has also shown the safety of ketorolac specifically following autologous BR. The routine addition of ketorolac following surgery has been shown to reduce patients' narcotic requirements in other fields of surgery, such as orthopaedics and gynaecology.

Given the abovementioned advantages of ketorolac, and the particular propensity for abdominally based free flap BR patients to be negatively affected by narcotic side effects, we hypothesize that the addition of ketorolac to a routine multimodal postoperative analgesia regimen will reduce hospital stay by a day or more.

The comparator for this study will be a placebo. Patients randomized both to the control group and those randomized to the intervention group will receive standard of care postoperatively, which will include pain control with scheduled acetaminophen and pro re nata oral and subcutaneous hydromorphone. In addition, control patients will receive a placebo with the same scheduling as the study drug, ketorolac. The rationale for this choice is to limit potential sources of bias, such as the patient and her care team, and maximize the accuracy of the primary outcome and patient-completed questionnaires that will serve as a secondary outcome. Other authors have previously successfully used placebos for ketorolac in their study design.

The majority of surgeons at our institution currently do not employ ketorolac in their postoperative pain control regimen. Thus, there exists clinical equipoise for patients considering participating and potentially being randomized into the control group, when compared with declining participation and receiving usual care.

This study will be single-centered and conducted in Nova Scotia, Canada. The Halifax Infirmary is a level III trauma center with multiple board certified Plastic surgeons who have undertaken extra training in microsurgery.

Participants will be removed from the study upon request. Her data up until that point will be used for statistical analysis, however she would no longer receive the intervention. Modifications to intervention dosing would occur if a participant develops any drug hypersensitivity reaction. Additionally any major harm attributed to ketorolac will prompt a termination of the intervention.

Each patient will undergo uni- or bilateral abdominally based free flap harvest, microsurgical anastomoses, and flap inset. Following randomization, the study drug will be sent to the main operating theatre, and attached to the patient's medical chart. At the conclusion of surgery, each patient will be administered their first dose of the study drug. Its content will depend on the patient's randomized group. If randomized to the ketorolac intervention group, the IV minibag will contain 30 mg of ketorolac tromethamine, or if in the control group it will be a sham normal saline IV minibag. They then will continue to receive their assigned drug every 6 hours for 72 hours. Patients all receive 1:1 care from nursing staff in the early postoperative period, and their nurse will administer all drugs given on the surgical ward.

All patients who have enrolled will receive the usual care for the postoperative period. The study drug will be in addition to usual flap monitoring, patient vital sign monitoring, progression of mobilization, drain management, along with the other care provided for these patients. Patients also will have the same background analgesic and anti-emetics medications ordered postoperatively, which is included on a standardized pre-printed order form at the Halifax Infirmary for this particular surgery.

Potential participants will first be introduced to the study following their initial consultation for BR. If the patient consents for abdominally based microsurgical BR, then the study will be reviewed, allowing time for any questions to be asked, and the risks and benefits to be discussed. The patient will be sent home with a study package that includes a copy of the study consent form and protocol. The study will be reviewed again at a preoperative appointment, allowing time for any follow up questions to be asked. This is a routine appointment intended for markings and patient questions regarding surgery, and the patient will officially be enrolled in the study if she wishes to participate at either time-point - the initial consultation or the preoperative appointment. Following the surgery, each patient will remain active in the study until her discharge from hospital. No follow up appointments or data will be scheduled or recorded for the purposes of the study (although follow up will occur as per usual with these patients and their respective surgeon). Therefore total enrolment time for each patient will be from the appointment immediately before surgery to her discharge from hospital.

A computer-generated algorithm will be used to randomize patients in a 1:1 allocation into the intervention or control group during surgery. Randomization will be stratified by laterality of the reconstruction (i.e. bi- or unilateral) because this could affect participant pain levels. Randomization will be managed and maintained by a third party service, not otherwise involved in the study, in the hospital pharmacy department. Allocation methods will not be disclosed to study investigators. Following randomization, the study drug will be sent to the main operating theatre, and given at the conclusion of the surgery. Both the intervention and control drug will be of the same storage and opacity to prevent identification by the patient or care providers. From there, the patient will then be administered the same drug according to the randomization result, based on delivery by the third party pharmacy service.

Randomization will be carried out within the pharmacy department placing patients into either the intervention or control group. The intervention and control intravenous drugs will be prepared in the same style IV minibag with the same volume and opacity of fluid. The fluids will be of similar viscosity. Once similar appearances and features are confirmed the prepared drug will be sent to the operating theatre or the ward as required depending on the time point.

During the conduction of this study all involved trial participants, clinicians, data collectors, event adjudicators, and data analysts will be blinded to the grouping of patients. All study medications will be stored in the hospital pharmacy department by a third party. Following randomization, treatment drugs will be dispensed by the pharmacy and then administered by nursing staff. The third party service will maintain records linking a patient unique identifier to her study group. Post hoc analysis to assess whether staff could identify between the intervention and placebo will be conducted.

Harms will be classified as either minor or major. Minor harms will include surgical site infection and those listed from prior clinical trials on the drug monograph, namely:

Common Clinical Trial Adverse Drug Reactions (>1%): Dyspepsia, abdominal pain, nausea, constipation, diarrhea, flatulence, gastrointestinal fullness, peptic ulcers, headache, dizziness, somnolence, edema

Less Common Clinical Trial Adverse Drug Reactions (≤ 1%): Eructation, stomatitis, vomiting, anorexia, duodenal ulcer, gastritis, increased appetite, melena, mouth ulceration, rectal bleeding, sore mouth, abnormal dreams, anxiety, depression, dry mouth, insomnia, nervousness, paresthesia, tinnitus, taste perversion, abnormal vision, blurred vision, deafness, lacrimation disorder, weight gain, alkaline phosphatase increase, BUN increased, excessive thirst, generalized edema, hyperuricemia, pruritus, rash, burning sensation, asthenia, pain, back pain, face edema, hernia, arthralgia, myalgia, joint disorder, chest pain, chest pain substernal, migraine, dyspnea, asthma, epistaxis, hematuria, increased urinary frequency, oliguria, polyuria, anemia, purpura

No major harms are anticipated. Major harms will include but are not limited to (* indicates possible association with oral ketorolac from drug monograph, <1%):

Surgical site hemorrhage requiring re-exploration in the main operating theatre Gastrointestinal hemorrhage* Pulmonary embolus Death

Any harm identified by care providers during enrolment of the study will be communicated to the principal and sub investigators. Minor harms will be communicated within 5 days. Any major harm will be communicated to the research ethics board within 24 hours. Complications will be managed medically in standard fashion, and the research will have no bearing on patient care. Emergency unblinding will be available to the patients' care providers at all times, as above.

Inclusion Criteria:

• Patients 18 years of age or older will be offered inclusion if they are scheduled to undergo unilateral or bilateral postmastectomy abdominally based free flap BR postmastectomy, either in a delayed fashion or at the time of initial mastectomy(ies).

Exclusion Criteria:

• Patients will be excluded if they meet any of the exclusion criteria: male; revision surgery; a planned pedicled flap reconstruction; any previous diagnosis of chronic renal disease, gastric ulcers, or bleeding disorder; acetaminophen, NSAID or specific ketorolac sensitivity or allergy; pregnancy; or opioid tolerance defined as preoperative opioid use greater than 50 mg.