Aim 1: To determine the effectiveness of perioperatively administered ketamine to decrease acute and persistent postmastectomy pain (PPMP).

Hypothesis 1.1: Patients undergoing partial or total mastectomy treated with a bolus and perioperative infusion of the NMDA-receptor antagonist ketamine will have decreased postoperative pain and opioid utilization compared to those receiving saline control.

Hypothesis 1.2: Patients undergoing partial or total mastectomy treated with a bolus and perioperative infusion of the NMDA-receptor antagonist ketamine will have decreased persistent postoperative pain measured at one year after surgery.

Aim 2: To determine whether there is increased power to detect therapeutic effectiveness in an interventional preventive trial, by enrichment with patients at high risk of PPMP.

Hypothesis 2.1: Ketamine will have a greater analgesic and opioid sparing effect on pain scores in high-risk patients than non-high risk patients, compared to placebo.

Hypothesis 2.2: Ketamine will have a greater preventive effect on pain burden scores at one year after surgery in high-risk patients than non-high risk patients, compared to placebo.

V. STUDY PROCEDURES

1. Preoperative evaluation

1. Psychosocial Assessment: After the patient gives informed consent, preoperative demographic, medical, psychosocial and pain questionnaires will be completed by the patient during their preoperative office visit. Questionnaires assessing psychosocial factors have been chosen based upon previous retrospective association with PPMP, and preliminary evidence from our prospective study in the same population, strong psychometric validation characteristics, and brevity. The Pain Catastrophizing Scale (PCS), which has been validated in pain patients and controls, will be used to measure catastrophic thinking associated with pain. Depressive symptoms, anxiety and sleep disturbance will be assessed using short-form instruments from the NIH roadmap initiative, Patient Reported Outcome Measurement Information System (PROMIS). The PROMIS instruments have been extensively validated in studies comparing results with established scales, and have been calibrated on over 20,000 persons.The Brief Symptom Index 18-Somatization Scale, also previously validated in chronic pain patients, will be used to measure somatization. The Positive Affect/Negative Affect Scale (PANAS) measures affective stance.

2. Pain assessment: The primary pain outcome measure will be Pain Burden Index as measured by the Breast Cancer Pain Questionnaire (BCPQ), which was originally developed by Gartner et al and used in over 600 patients in our previous studies. The BCPQ includes assessment of average pain, pain frequency (how many days/week), and specific assessment of pain at 4 body areas (breast, arm, side, axilla), from which a Pain Burden Index (PBI) can be calculated. A later iteration of the BCPQ also includes questions about other body pain, seeking medical help for pain, painkiller use, neuropathic/sensory symptoms, and functional impairment due to pain, and has been successfully employed in our current prospective observational study. We will also use the Brief Pain Inventory, a well-validated general measure of pain and disability worst pain, least pain, and interference. We will supplement this with a standard VAS assessing pain with and without arm movement and deep inspiration. The Pain Coping Questionnaire will be administered to prospectively assay for effective coping strategies used by patients and we will use a 2 item version of this questionnaire to assess the use of these strategies.

3. Quantitative sensory testing:

   1. Pressure pain testing methods: Pressure pain threshold and tolerance will be assessed in a similar manner to our previous studies using a digital pressure algometer (Wagner FDX, Greenwich, CT, USA) with a flat round transducer, probe area 0.785 cm2. Testing sites will be bilateral on the dorsal aspect of the proximal forearm (extremity site) and over the trapezius muscle at the upper back (truncal site). For pressure threshold determination, pressure is increased at a steady rate of approximately 1 lb/second until subject indicates that the pressure is first perceived as painful. Pressure Pain Tolerance will be determined by instructing patients to indicate the pressure at which the pain was no longer tolerable. Two trials will be performed at each site.
   2. Temporal Summation testing methods: Mechanical pinprick pain will be assessed in a similar manner to our previous studies using standardized weighted pinprick applicators similar to those described by Rolke et al using a range of forces (128 mN, 256mN and 512mN) which result in a painful sensation in most subjects. First, a single stimulation of the lower force pinprick will be applied to the dorsal aspect of the index finger between the first and second interphalangeal joints of the each hand while resting palm down on the armrest, and then rated by the subject on a scale of 0-10. Next, after a break of at least 10 seconds, a train of 10 stimuli will be applied at the same premarked spot, at a rate of 1 stimulation/second. The subject will rate pain on a scale of 0-10 after the first, fifth and tenth stimulus, then rate any ongoing pain 15 seconds after cessation of the last stimulus. The same procedure will be repeated on the third finger of the each hand. Subjects will undergo two trials at each force, with a break of at least one minute between trials.

   2. Perioperative intervention Patients will receive a 0.5 mg/kg dose of IV ketamine (max dose 50 mg) or the same volume of saline 15 minutes after induction of general anesthesia. Thereafter, a continuous infusion of 0.15 mg/kg/hr ketamine with a maximum dose of 15 mg/hr or saline will be administered until the end of surgery (when the surgical closure has begun, approximately 30-60 minutes prior to emergence from general anesthesia). The investigational drug service will prepare blinded solutions (2 mg/mL ketamine or saline) and deliver to OR pharmacy before the beginning of each case to hands-on providers. Cost of ketamine is $12 per 200 mg vial, and will be covered by the NIH K23 budget. General anesthesia will be provided by the hands-on anesthetist, who will be free to use at their discretion the following anesthetic drugs, which are typically used in this population, and are all compatible in solution with ketamine: Midazolam 2 mg, Fentanyl <500 mcg, Dilaudid <5 mg total, Propofol 1-2 mg/kg, Lidocaine 50-100 mg, Vecuronium, Rocuronium, succinylcholine or Cisatracurium as needed, Zofran 4 mg, Decadron 4 mg, Cefazolin, Glycopyrrolate and Neostigmine as needed for reversal, as well as inhalational agents other than nitrous oxide.

Subject and investigator blinding of screening and treatment: As part of informed consent, subjects will be appraised of the fact that the study investigates risk of short- and long-term pain and whether people in different risk groups respond to a treatment, but will not be told whether they are considered high- or low- risk, or whether they received ketamine or placebo. Effectiveness of blinding will be assessed by querying patients about which group they think they are in on POD1. Investigator and hands-on provider will be blind to treatment (ketamine vs placebo). Nursing staff in PACU will be appraised of the fact that patients are in a study, but also blinded to the treatment. Effectiveness of blinding of study staff who is asking patient to rate perioperative pain will be achieved by having this study staff record their impression of group on a single item questionnaire in the Redcap system. Responses of patient and study staff as to their impression of group will be compared to actual group at the end of the study.

3. Perioperative data collection

1. Surgical, Anesthetic, and analgesic data collection: Using the Weiner preoperative evaluation center email alert system, the hands-on anesthetist and/or anesthesiologist will be contacted to inform them on the patient's participation in the study, and given instructions about the volume of bolus and rate of infusion. The total time and volume of infusion will be recorded will be recorded in the anesthetic record, so that a total dose of ketamine can be calculated. In order to adjust for potential impact of other anesthetic techniques or pharmacological agents on PPMP, the following data will be extracted from the electronic anesthesia record: Primary anesthetic method (volatile vs. IV propanol vs. regional/paravertebral vs. combined PVB/GA), amount and type of local anesthetic administered, and opioid analgesic used (the latter expressed as morphine equivalent units/hr). In addition, information about other relevant analgesic preoperative, intraoperative, and postoperative medications such as Oxycontin, gabapentin, pregabalin, and dexmedetomidine will be gathered from the electronic record. Other potentially confounding intraoperative factors such as vasopressor use/amount, fluids or blood administered, total anesthesia time, and airway management (LMA, ETT, or native) will be recorded. Among this group of breast surgeons, there is relative homogeneity in anatomical technique and tissue disturbance, although there is some variation in dissection technique. Detailed assessment of surgical technique will include primary surgeon, duration of surgery, presence of bilateral surgery, type of surgery (total mastectomy, segmental mastectomy), performance of nodal dissection (axillary or sentinel node), and presence and type of concurrent reconstruction.

2. Acute postoperative pain assessment: Initial pain score upon arrival will be assessed by PACU nursing staff and/or anesthetist. One hour and 24 hours after PACU arrival, patient will be assessed by research study personnel, being asked to rate pain at rest, with arm movement, and with maximal deep inspiration using the pain verbal rating scale (VRS). Presence of nausea and treatment with supplemental antiemetics in recovery area will be recorded. In addition, opioid consumption and adjuvant analgesic use will be ascertained from the patient's medication administration record. Upon discharge, which is typically on POD1 or 2 for mastectomy patients, total postoperative opioid use/hours of postoperative time as inpatient will be calculated and expressed in morphine equivalent units/hr. On postoperative day 1, mastectomy patients will undergo brief bedside QST (as at preoperative evaluation, described above).

   3. Persistent pain, long term psychosocial and psychophysical assessment: Follow up assessments of persistent pain using the BCPQ, BPI, VAS, and pain coping questionnaires will be made at 2 weeks, 3 mo, 6 mo, 12, and 24 mo, with follow-up either via secure link to the REDcap electronic data capture system. Patients are seen at visits with the surgical team, or oncologists at these times. The brief QST and psychosocial questionnaires will be re-administered at 2 weeks and 1 year postoperatively during one of the patient's follow-up visits.

Inclusion criteria:

• female
• aged 18-85 years
• scheduled for total or partial mastectomy
• willingness to undergo psychophysical and psychosocial testing
• willingness to participate in long-term follow-up
• willingness to be randomized to treatment with IV ketamine or saline during general anesthesia.

Exclusion criteria:

• scheduled for biopsy only
• pregnant
• elevated ICP
• schizophrenia or bipolar disorder
• allergy to ketamine
• Class III or higher heart failure

women undergoing breast surgery

Investigational Drug service will prepare ketamine vs placebo infusion for administration during surgery