Cancer pain is one of the most frequent and relevant symptoms in cancer patients and has a great impact on a patient's quality of life. International and local standards recommend as an initial strategy, the use of an analgesic scheme composed of strong opioids (morphine, methadone or fentanyl) associated with adjuvants such as paracetamol, based upon the assumption that the use of combined analgesics could have a better analgesic effect, could allow the use of lower dose of opioids and that also could prevent the occurrence of adverse effects of opioids. However, there is uncertainty about the impact of paracetamol as an adjuvant in patients who use strong opioids for pain management in cancer patients with moderate to severe pain.

To clarify this question, this study aims to evaluate the efficacy and safety of intravenous paracetamol associated with strong opioids in hospitalized cancer patients who have pain associated with cancer of moderate to severe intensity, (4 or more), older than 18 years.

Randomized double-blinded controlled study comparing intravenous acetaminophen 1 gr 4 times a day versus placebo for 48 hours as an adjuvant to strong opioids. We will assess pain intensity as a primary outcome validated assessments that estimate Verbal Numerical Rating Scale (VNRS) analogous verbal pain from 0 to 10, and de visual Analog Scale (VAS). We estimated that a decrease of 1 point on the verbal numerical scale would be statistically significant. In addition, the investigators will calculate the amount of total opioid dose in 24 hours and then perform the intervention. As a secondary outcome, adverse effects such as drowsiness, constipation, nausea and vomiting would be evaluated

This is a randomized, controlled, double blind, parallel-group, single center clinical trial. This study received ethical approval by the Ethics Committee of the Pontificia Universidad Católica de Chile (ID #180328004). The study protocol was designed using the recommendations of the Consolidated Standards of Reporting Trials (CONSORT) statement.

The setting of this study is the General Internal Medicine Ward of a tertiary level university hospital (UC Christus Clinical Hospital) where patients will be recruited.

The hypothesis of this study is that in hospitalized oncology patients over 18 years of age, acute pain management with strong opioids plus intravenous paracetamol is not superior to the use of strong opioids alone.

Participants Patients 18 years old or older, diagnosed with cancer and admitted to UC Christus Clinical Hospital of any ethnicity or nationality with moderate to severe pain.All participate in the study will sign an informed consent form.

Treatments Opioid administration - Before starting the study a standardized pain management protocol for cancer patients with moderate to severe pain will be implemented across the institution. In this protocol, a standard analgesic protocol (scheduled strong opioids plus rescue doses, such as morphine, methadone or fentanyl) will be started by the ward team upon admission in order to ensure that all cancer patients with moderate to severe pain will have an adequate analgesic scheme for pain control regardless study enrollment. Briefly, cancer patients with moderate to severe pain who are opioid naive will be started on scheduled morphine, methadone or fentanyl by continuous infusion plus rescue doses. Standard doses will be recommended but these doses could be changed by treating clinicians according to clinical judgement. For patients with prior use of opioids, ward or treating clinicians could start scheduled methadone or morphine or fentanyl continuous infusion increasing the prior opioid dose. Early consultation to palliative care clinicians will be recommended for this population.

Acetaminophen - experimental group - Intravenous acetaminophen is usually delivered in a 100cc solution that is prepared in a transparent glass bottle. As the placebo cannot be prepared in the same type of bottle, the content of the acetaminophen preparation will be transferred to a standard 100cc plastic flask for IV infusions, which will be labeled with the name and ID number of the patient, with the drug to be administered labeled acetaminophen/placebo (including both names) and with the allocation number for the randomization. The preparation of the acetaminophen plastic flask will be indistinguishable from the placebo plastic flask. As the intervention will last 48 hours, 8 plastic flasks with the drug will be sent directly to the clinical nurse in charge of the administration of the drug in the general ward.

Placebo - control group - The placebo will be prepared using the same plastic flask as used in the acetaminophen group. In the placebo group it will be filled with 100cc of saline, and will have the same label as the acetaminophen group, therefore they will be indistinguishable from each other. In the case of the placebo group, 8 plastic flasks with the placebo will be sent directly to the clinical nurse in charge of the administration of the drug in the general ward.

Other treatments - As the research group wants to assess the impact of this intervention in the real clinical setting, treating clinicians will be allowed to be added according to clinical judgement. Non-steroidal anti inflammatories (NSAIDS), steroids, anticonvulsants, or other adjuvants could be added.

Randomization - After informed consent, eligible patients will be randomized into two arms: (A) acetaminophen or (B) placebo. The randomization procedure will be performed by the institution's pharmacist using a web-based randomization software platform specifically designed to support data collection for research studies (Research Electronic Data Capture, REDCap®), platform that provides automated export procedures for data downloads. The randomization will be performed following a stratified block randomization, with blocks of 4 or 6 patients among which 50% of each block will receive placebo and 50% will receive acetaminophen. The study will be blind with randomization concealment. Once the pharmacist has identified the allocated arm of the enrolled patient, a total of 8 identical plastic flasks will be prepared in the pharmacy service, with a total amount of 100ml of volume each and each one labeled as previously described. For arm (A), IV acetaminophen will be prepared, for arm (B), IV saline will be prepared. The eight plastic flasks will be delivered to the general ward and the clinical nurses will be in charge of administering the infusions during the 48 hours study period. Precautions will be taken to ensure that treating physicians, clinical nurses, data collectors, data adjudicators, patients and researchers will be blind to patient allocation.

Follow-up and data collection - Patients will be asked to complete a baseline assessment questionnaire and then two other questionnaires at 24 and 48 hours after enrollment. The questionnaires were selected seeking to assess the primary and secondary outcomes and variables that could impact patients' pain experience. We included a variety of questionnaires and instruments, to assess eligibility criteria, the primary and secondary outcomes and possible affect modifiers, including instruments to assess delirium (Memorial Delirium [MDAS]), pain (Verbal Numerical Rating Scale [VNRS] and Visual Analog Scale [VAS]), use of analgesia prior to admission, alcohol and drug consumption, risk of chemical coping, symptoms, psychological distress (Hospital Anxiety Depression Scale [HADS]), quality of life and symptoms associated with their current hospitalization .

Data collection and management Sample size - To estimate the sample size, we decided to use the strategy of identifying the minimum clinically important difference (MCID) in pain according to the ESAS scale, which evaluates pain on a scale from 0 to 10, similar to the VNRS, which we will use as our main outcome. The MCID is defined as "the smallest change in a measurement that signifies an important difference in a patient's symptoms". In a study conducted by Farrar, et al, the MCID for pain was defined as 2 points, which was evaluated in a short in-hospital follow-up period, a scenario that is similar to that of our study. In another study, conducted by Hui, et al , published in 2015, different methods for establishing MCID were evaluated. In that study, using the anchor-based method through the calculation of the ROC curve, it was recognized that an improvement in pain intensity by 1 point on the ESAS scale was identified by patients as a clinically significant improvement, i.e. patients detect 1 point on the ESAS scale as an improvement in pain control, scale similar to the VNRS. In this study the standard deviation for the pain score was 3 points, similarly to what was found in previous studies. Using other similar strategies, a difference between 1 and 2 points was identified as clinically significant. However, in this study the pain assessment was performed on an outpatient basis and with a 3-week difference between the first and the last assessment.

In an unpublished sample of 100 advanced cancer patients assessed in our PC unit, we found that the mean intensity of pain using the VNRS among patients with moderate to severe pain was 5,8 points with a standard deviation (SD) of 1.7 points.

From the data obtained from prior publications, considering an alpha of 0.025, with a power of 0.8, we estimated that a sample size 112 patients would be required, with 56 patients in each group to detect a difference of 1 point in pain intensity between the groups, with a standard deviation (SD) of 1.7. These assumptions are supported by the following reasons:

Because a difference of 1 point is considered to be what is clinically defined as significant, so we should try to detect a difference greater than that.

Because we reported an SD of 1.7 in the initial pain scale in a sample of cancer patients in our unit.

In this way the investigators could be able to recommend, not to use intravenous paracetamol, which would generate an indirect recommendation not to use oral paracetamol in patients with moderate to severe acute pain associated with cancer. This not only has an economic impact, but also could affect the well-being of patients who are sometimes in a great pharmacological burden, in the context of low oral intake and frequent nausea and vomiting.

Inclusion Criteria:

• Cancer patients hospitalized at UC Christus Clinical Hospital of any ethnicity or nationality.
• With acute pain> or = a 4 in Verbal Numerical Rating Scale (VNRS)
• They can be patients who are virgins to opioids or previous users of weak or strong opioids.
• They may have somatic, visceral or neuropathic pain
• They may be users of NSAIDs or corticosteroids

Exclusion Criteria:

• Patients who refuse to enter the study
• Patients who don´t speak Spanish a mother language
• Patients who present a qualitative or quantitative awareness commitment that prevents the assessment of pain.
• Patients with acute liver failure or chronic liver damage Child C.
• Patients allergic or hypersensitive to paracetamol.
• Patients with a prognosis of life less than 72 hours (evaluated according to clinical criteria)

For numerical variables, baseline characteristic data were analyzed and represented with mean±SD, or median with interquartile range (IQR), according to data type and distribution. Categorical variables are presented as frequency and percentage.

For bivariate analysis, Student T-test or ANOVA were used for enough sample size to assume Normal distribution of the averages by Central Limit Theorem, and Wilcoxon's rank sum test for non-normal distribution data.

Finally, multivariate mixed models will be adjusted for interest scores: VAS, VNRS and MEEDs at 48 hours. As fixed effects in the model, we will consider de intervention (placebo or paracetamol) and other factors (as tabacco) and the patient as the random effect. A significance of 0.05 will be considered. The analyzes will be performed using the statistical package STATA 14.0.

We are unable to make the data available in a public repository or uploaded as supplementary information because this is not permitted by our organisation's research governance policy and Ethics committee regulations. Anonymised data can be made available to researchers who meet the conditions of the ethics approval and research governance policy that applies to this study. Researchers may request anonymized data access by contacting Universidad Católica's School of Medicine Research Office (DIDEMUC) at didemuc@med.puc.cl.

matching placebo. prepared by pharmacist