Tonsillectomy is one of the most common surgical procedures in the United States, and post-operative pain management is a challenge for otolaryngologists. A 2013 black-box warning on codeine following tonsillectomy has drawn attention to potential concerns with all narcotics in these patients, and many surgeons try to avoid narcotics. The use of intra-operative corticosteroids has been proven to be beneficial in reducing post-operative morbidity, and some small studies have shown possible benefit to the use of post-operative oral corticosteroids as well, although the results of these studies are mixed. To date, no one has looked at whether the use of post-operative oral steroids may reduce or eliminate the need for narcotics. We aim to determine whether the addition of oral steroids to our post-operative pain regimen can reduce the need for narcotic pain medications.

I. Background

Tonsillectomy is one of the most common surgical procedures in the United States with the predominant indications being obstructive sleep apnea and recurrent tonsillitis1. While it is usually performed on young and otherwise healthy children, postoperative morbidity is ubiquitous occurring in nearly 20% of children with the most frequent complaints being throat pain, otalgia, fever, poor oral intake, nausea and emesis, and secondary hemorrhage2. Historically, Tylenol with codeine was commonly used for pain control post-operatively in children. However, in 2013 the FDA issued a black box warning on codeine following tonsillectomy due to reports of multiple deaths and morbidities in children3. Since that time, there has been increased awareness of the dangers of opioids in general and many surgeons are trying to minimize or eliminate the prescription of opioids to children following tonsillectomy.

It has been found by multiple studies that intra-operative use of steroids during tonsillectomy has beneficial effects in children. A Cochrane Review updated in 2011 presented a meta-analysis of 19 randomized, placebo-controlled, double blinded studies and concluded that routine use of a single intra-operative dose of dexamethasone reduces post-operative morbidity. The studies found a decrease in post-operative emesis and a higher likelihood of advancement to a soft or solid diet on the first postoperative day. Eight of the studies used the Visual Analog Scale (ranging from 0-10), which found a reduction of post-operative pain of more than one point. No adverse effects related to intra-operative corticosteroid use were reported and a single dose of steroids is believed to be virtually harmless4. Despite the utility of corticosteroids intra-operatively, there is currently a paucity of studies investigating the potential effects of a short post-operative course of oral (PO) steroids on pain control and other morbidities of tonsillectomy. Of the available literature, a short course of prednisolone of up to 1 week in patients without specific contraindications was found to provide a significant benefit in pain relief, return to normal diet, return to normal activity, and re-epithelialization of the tonsillar fossae, in an assessment of 69 pediatric patients5. Another study by Macassey et al. compared 5 days of oral prednisolone to placebo by assessing pain, nausea, vomiting, return to normal function, and sleep duration and quality. This study did not find a difference in any of the study parameters between the two groups6. To our knowledge, there have been no studies specifically looking at how the addition of a post-operative course of oral steroids impacts patient's necessity for narcotic pain medications.

Due to conflicting reports and a general paucity of literature to support or refute routine postoperative steroid use in children following tonsillectomy, it will be useful to determine the efficacy of this otherwise ubiquitous and benign medication on pain control, time to return to normal diet, dehydration, post-operative infections, and post-operative bleeding in our patient cohort; these are the most common complaints and complications following tonsillectomy surgery. Therefore, similar to the above-referenced studies5,6 we will be evaluating all of these endpoints in our patients. However, in addition to these endpoints we will also look at the need for post-operative narcotics. Given the recent black box warning on Tylenol with codeine use in children following tonsillectomy3, many surgeons are understandably hesitant to prescribe any narcotics to children following tonsillectomy. Our study will evaluate the potential role of steroids to reduce or possibly even eliminate the need to prescribe these children narcotics.

II. Objectives

We propose a double-blind, randomized placebo-controlled study comparing the post-tonsillectomy morbidity and analgesic/narcotic use in children 4 to 18 years old receiving a 5 day-course of oral prednisolone versus those receiving placebo. We plan to approach patients and parents preoperatively to assess their willingness to participate in the study. Following tonsillectomy, parents will be asked to complete a patient log, detailing the child's post-operative course. This log will document the quantity of narcotics and other pain medications used, daily pain assessments using the validated Wong-Baker facies tool, return to regular oral diet, and return to regular activities. Complications such as dehydration, post-operative infection, and post-operative bleeding requiring an emergency room visit will also be assessed in a post-operative survey, which will be completed at the post-operative office visit.

Because the use of a single dose of intraoperative steroids is a widely accepted practice, we believe there is a value in investigating whether a short postoperative course of steroids adds further benefit to a child's post-tonsillectomy course, especially because it is a safe and inexpensive intervention, and may reduce the need for post-operative narcotics. Given the significant concerns with narcotics in this cohort of patients and the recent black-box warning, there is great interest among pediatric otolaryngologists in minimizing narcotic use. This study has significant potential to change the paradigm of post-operative pain management after tonsillectomy to minimize or even avoid the use of narcotics.

The primary endpoints for this study include frequency and total amount of narcotic and non-narcotic pain medication required, time to return to normal diet, and time to return to normal activity. Secondary endpoints will include the rates of post-operative emergency department visits, dehydration, and post-operative tonsillar hemorrhage.

III. Methods

We propose a prospective double-blinded randomized placebo-controlled control trial to assess the effects of oral prednisolone on the post-operative course following tonsillectomy.

Recruitment At the pre-operative visit, a member of the patient care team (resident, nurse practitioner, or attending physician within the Pediatric Otolaryngology department) will assess the willingness of the patient and parents to hear more about the study. Study details will be made available to the parents and child, with the understanding that their child will be randomized into a group receiving post-operative steroids or a group who receives placebo. It will be emphasized that study participation is voluntary and without participation children will receive our current standard of care, which is a single intra-operative dose of dexamethasone but no post-operative oral steroids. It will be made clear that the parents are able to withdraw from the study at any time. If they agree to take part in the study, parents will be asked to record all pain medications taken by their child in the five days following surgery as well as keep a record of diet and activity. Informed consent will be obtained for participation in the study at the first visit. Parents will also be asked to bring the study medication bottle and the pain medication bottle to the follow-up visit to ensure compliance. A survey assessing secondary endpoints will be completed at the post-operative follow-up visit. Study details and patient logs will be available in English only.

Subjects Pediatric patients aged 4-18 undergoing tonsillectomy at the Children's Hospital of Michigan by Dr. Bianca Siegel, Dr. Nathan Gonik, or Dr. Mallory O'Niel will be eligible to participate in this study.

Inclusion criteria -

1. Subject is male or female age 4-18 years of age
2. Patients undergoing total tonsillectomy with or without adenoidectomy or tympanostomy tubes
3. Patients with obstructive sleep apnea or recurrent tonsillitis requiring tonsillectomy.
4. The patient or caregiver is able to provide written informed consent.
5. Patients with American Society of Anesthesiologist (ASA) physical status 1 or 2.

Exclusion criteria -

1. Subjects with complex medical conditions or craniofacial abnormalities
2. Subjects with known personal or family history of bleeding disorder
3. Subjects with cognitive or developmental disorders
4. Subjects currently taking corticosteroids for other medical conditions, or who have taken corticosteroids within 2 weeks of surgery
5. Subjects who are wards of the state
6. Subjects who have had an organ transplant
7. Subjects who are on other immunosuppressant medications
8. Subjects with diabetes mellitus
9. Non-English Speakers or English as a second language
10. Subjects undergoing intracapsular tonsillectomy
11. Subjects with American Society of Anesthesiologist (ASA) physical status greater than 2.
12. Pregnancy

Pre-operative Visit (Visit 1) If the patient and family agree to participate after an explanation of the project (see Recruitment above), and meets all criteria for participation, they will be enrolled in the study. An informed consent document, oral assent document (if applicable), and written assent document (if applicable) will then be obtained and a copy will be given to the patient and/or parent. Only one parent will be required to provide consent for the surgery and for participation in the study as typically only one parent accompanies children to our clinic. A physical examination will be performed to note the patient's tonsil size. Pre-operative sleep study findings in patients with suspected obstructive sleep apnea will be recorded as well. Any concurrent medications will be recorded. The patient's information will be sent to the participating pharmacist for randomization into an experimental arm (5 day course of post-operative oral steroids - prednisolone) and a control arm (5 day course of post-operative ORA-Sweet © placebo) Randomization will be performed in a 1:1 ratio at Visit 1 by the participating pharmacist, using www.randomization.com. The patients as well as the surgeons and the remainder of the research team will be blinded to the randomization. Prescription documents for the study medication and placebo will be created by the participating pharmacist prior to the start of the study. Once a patient is enrolled, a prescription will be sent to the participating pharmacist who will then randomize the patient to receive either prednisolone or the equivalent volume of OraSweet © solution to serve as a placebo. Details of randomization as well as which patients received the study medication will be kept by the participating pharmacist and can be available for review by the institutional review board or any governing agencies at their request.

Day of Surgery (Visit 2) On the day of surgery all patients will receive a single intravenous, weight-based dose of dexamethasone intra-operatively, as this is the current standard of care. Tonsil size will, once again be noted at this visit and updates to concurrent medications will be recorded. Any adverse events intraoperatively or immediately post-operatively will be recorded. In female patients over age 9, it is hospital policy that the patient undergoes a urine pregnancy test prior to undergoing surgery. This will be recorded and, if positive, the patient will not undergo surgery and will be disqualified from the study. After surgery, the participating pharmacist will then deliver the appropriate volume of study medication (1mg/kg/day prednisolone divided BID (twice per day)) or for 5 days post-operatively to start on post-operative day 1. The control group will receive the same volume of OraSweet© placebo also to start on post-operative day 1. Each group will receive an additional quantity of 3 doses in the event of emesis and a need to redose. Parents will then be asked to monitor their child's post-operative pain, oral intake, activity, and quantity of pain medication given for 5 days using patient logs created for this research endeavor (attached). The patient will then be discharged home or admitted for overnight observation if pre-operative sleep study scores indicate a severe obstructive sleep apnea (AHI >10) based on recommendations by the American Academy of Otolaryngology - Head and Neck Surgery Foundation or at the surgeon's discretion.

Post-operative visit (Visit 3) The first follow up visit will occur 2 weeks after surgery. At this visit, we will collect the patient logs used by the parents as well as the study medication bottles to ensure compliance or to determine continued eligibility in the study. At this visit, we will perform a physical examination to ensure healing of the wounds as well as record any adverse events and other, concurrent medications. Parents will also be asked to complete a brief survey, asking about the post-operative course and any complications (attached). We will then transfer the paper patient log recordings into an electronic format kept securely on the secure DMC server. Once statistical power has been reached, we will use these data to evaluate whether a 5-day course of post-operative steroids improves post-operative pain and decreases time to normal activity and oral intake. If patients develop any postoperative complications, the standard of care for that complication will be undertaken.

Medication Details On the day of surgery, medication will be prepared for the patients by the participating pharmacist, which the parents will receive in the recovery room. The medication bottle will either contain 15 mg/5 mL prednisolone elixir (Prednisolone group) or ORA-Sweet © Syrup (Placebo group). The medication will be provided in a marked bottle by the pharmacy. The medication will be supplied specifically for this study and will be stored at room temperature in the investigational drug room in the inpatient pharmacy until needed. Medications will be ordered specifically for this study and will have its own medication lot independent from the regular hospital supply. A weight-based dose of 1 mg/kg/day of prednisolone elixir rounded to the nearest whole number and divided BID (twice per day ) for 5 days will be given with a maximal dose of 60 mg per day for the intervention group. This dosing schedule is similar to other common pediatric uses of prednisolone including acute asthma attacks. The placebo group will take the equivalent volume of OraSweet © Syrup placebo. An additional volume of 3 doses will be provided to ensure proper redosing in the event of emesis. Marked oral syringes will be provided by the pharmacy for ease of dosing. The study medication or placebo is to begin on post-operative day 1 and will continue twice daily through post-operative day 5. It will be suggested that all medications be taken with food to minimize gastrointestinal side effects. Other post-operative medications will be standardized and all will receive oxycodone elixir (0.05 mg/kg every 4 hours (Q4hr) as needed), acetaminophen (15 mg/kg every 4 hours (q4hr) with a maximum dose of 500mg per dose) and ibuprofen (10 mg/kg every 6 hours (Q6hr with a maximum of 400mg per dose) to be used regularly. No post-operative antibiotics will be given as this is currently recommended against by the American Academy of Otolaryngology - Head and Neck Surgery Foundation. Parents will be advised to use acetaminophen and ibuprofen as primary medications for pain control, and to use oxycodone only if inadequate pain relief with alternating acetaminophen and ibuprofen.

Sample Size Determination Statistical analysis will be performed to evaluate for any differences between the control and experimental groups using an analysis of covariance design. For a 2 group design with 2 covariates and a medium-to-large effect size (alpha 0.05), a sample of about 80-180 participants would be needed to achieve a power of 0.80 or above. Therefore in order to allow for some drop-out, our goal is to achieve a sample size of 200 participants.

Patient logs Attached are the patient logs that are to be used for this research study. Aside from the Wong-Baker Faces scale, the remainder of the items in the patient logs to be used are non-validated as no such tool currently exists for use of this purpose. Parents will be asked to fill out a daily log for 5 days post-operatively focusing on aspects of pain control, medication administration, activity level, and oral intake. Parents will be asked to evaluate the child's pain on each post-operative day. Pain will be assessed in all patients using the validated Wong-Baker FACES scale, which has been incorporated into our study patient logs. Parents will be given a packet with written instructions and space to record the child's pain level on each post-operative day, as well as space to record their pain medication administration. They will also be given an activity and diet diary to monitor the child's activity and diet.

Post-operative Survey Parents will be given a survey to complete at the post-operative visit (Visit 3). The survey will be completed while they are at the office, so that we can clarify or ask any further questions as needed. The survey will assess general compliance with the study, as well as ask about any complications, including bleeding, ED visits, or hospital admissions. This will allow us to capture any potential visits to outside emergency rooms which we would not otherwise be aware of. If the parents indicates "yes" to the question about whether they visited an emergency room, further details will be elucidated.

Data analysis Outcome measures will include: the amount of post-operative narcotics required in the 5 day post-operative period, amount of other non-narcotic analgesic medications (acetaminophen or ibuprofen) required in the 5 day post-operative period, time to return to normal activity, and time to return to normal diet. Complications such as post-operative bleeding, dehydration that requires an emergency department visit, and post-operative infection requiring antibiotics will be monitored as well. As study subjects with complications return to the Children's Hospital of Michigan Emergency Department, the study team will be notified by the on call otolaryngology resident and appropriate measures can be undertaken. Complication rates will be compared between the two study groups.

Drug Safety Monitoring Plan Adverse effects associated with post-operative oral steroids or complications from tonsillectomy will be monitored via the research team using the data collection sheets. During the period that the study is ongoing, Dr. Bianca Siegel will compile all information on adverse events every three months beginning after the first 20 patients. The pharmacy will reveal the randomization groups and statistical analysis will be performed to determine any significant difference in adverse events between the two groups. Statistical analysis will be performed by a Wayne State University biostatistician. We will monitor specifically for serious adverse events such as dehydration or post-operative tonsillar hemorrhage. If a statistically significant increase in serious adverse events is found between the study group and the placebo group, the study will be aborted. We will also abort the study if there is a non-statistically significant difference, but a noticeable trend of increased adverse events in the study group. Early in the study, our small sample size may result in a lack of statistical significance despite a large increase in relative risk of serious adverse events.

IV. Benefits

The potential physical benefits to a subject participating in this study include:

• Decreased narcotic use
• A faster return to normal activity
• A faster return to normal diet
• Decreased post-tonsillectomy complication rates
• Improvement in severity or resolution of obstructive sleep apnea
• Improvement in rates or resolution of recurrent tonsillitis/pharyngitis

The potential benefits to society include:

• Data-based information on the improvement of a normally very painful post-operative tonsillectomy course following administration of prednisolone, with decreased necessity for narcotic prescriptions
• The potential for a new standard of care for improving tonsillectomy surgical outcomes.
• Decreased costs to society from readmissions following tonsillectomy

V. Risks Short-term corticosteroid use is general considered safe. The most common adverse reactions to steroids in the pediatric population include gastrointestinal upset, hyperactivity, agitation, mood swings, and insomnia (up to 10%). Other, more infrequent risks include adrenal suppression, delayed wound healing, glucose intolerance, myopathy, neuritis, and weight gain (1-5%). These adverse events typically resolve with the discontinuation of the corticosteroid. There are some reports of avascular necrosis of the hip, although these are exceedingly rare. Corticosteroids, including prednisolone, are used for a variety of pulmonary, rheumatological, and musculoskeletal pathologies in children and have stood the test of time for safety and efficacy. These risks are mitigated by the short duration of use of the medication, the encouragement to take the medication with food, the recommended dosing schedule, and proper preoperative counseling and education regarding administration of prednisolone.

As a part of this study and the current standard of care following tonsillectomy at DMC Children's Hospital of Michigan, subjects will be prescribed acetaminophen (Tylenol ®) and ibuprofen (Motrin ® or Advil ®) if acetaminophen is not tolerated. Oxycodone will also be prescribed for additional pain coverage as needed as a normal part of the post-operative course. A list of the more common and serious risks of each of the medications is as follows:

• Acetaminophen: swelling, itching, liver failure (very uncommon)
• Ibuprofen: nausea, vomiting, swelling, ringing in the ears (3-9%)
• Oxycodone: dizziness, nausea, vomiting, fatigue, sedation

There are risks associated with undergoing tonsillectomy including, but not limited to, dehydration, bleeding, infection, trouble swallowing, painful swallowing, sore throat, temporary voice changes, difficulty breathing, and, very rarely, death. These risks can be lessened by good oral hydration and adequate pain management after surgery.

Social risks are unlikely to occur during the course of this study, but there is a small risk of breach of confidentiality. In order to maintain your child's confidentiality, all of their identifiable information will be kept separate from study documents.

Inclusion Criteria:

1. Subject is male or female age 4-18 years of age
2. Patients undergoing total tonsillectomy with or without adenoidectomy or tympanostomy tubes
3. Patients with obstructive sleep apnea or recurrent tonsillitis requiring tonsillectomy.
4. The patient or caregiver is able to provide written informed consent.
5. Patients with American Society of Anesthesiologist (ASA) physical status 1 or 2.

Exclusion Criteria:

1. Subjects with complex medical conditions or craniofacial abnormalities
2. Subjects with known personal or family history of bleeding disorder
3. Subjects with cognitive or developmental disorders
4. Subjects currently taking corticosteroids for other medical conditions, or who have taken corticosteroids within 2 weeks of surgery
5. Subjects who are wards of the state
6. Subjects who have had an organ transplant
7. Subjects who are on other immunosuppressant medications
8. Subjects with diabetes mellitus
9. Non-English Speakers or English as a second language
10. Subjects undergoing intracapsular tonsillectomy
11. Subjects with American Society of Anesthesiologist (ASA) physical status greater than 2.
12. Pregnancy