Off-label prescription of xed-dose combination of tramadol and dexketoprofen in primary health care. Evidence-based or cause for concern?

Background: The increase of consumption of the xed-dose combination of tramadol/dexketoprofen in Spain and other countries is noteworthy. The authorised therapeutic indication is symptomatic short-term treatment of moderate to severe acute pain in adult patients. Objective: Describe the pattern of use of tramadol/dexketoprofen in the eld of primary health care to examine potential off-label prescribing and warn about possible risks. Methods: A descriptive, cross-sectional and multicenter study carried out between March 2017 and March 2018. Total population were patients covered by the public health service in Catalonia, Spain, with an active prescription of tramadol/dexketoprofen on March 28, 2018. Target population were those patients who were prescribed tramadol/dexketoprofen over 20 day’s treatment. Results: There were 176 patients with active prescription of tramadol/dexketoprofen. All patients (100%) had a treatment duration exceeding 5 days and 72.7% (N=128) exceeding 20 days. The average length of treatment was 224±160.8 days. 35.1% of patients were treated with >2 medicines for pain concomitantly with tramadol/dexketoprofen. Conclusion: Tramadol/dexketoprofen xed-dose combination in practice was used frequently off labelled according to the product characteristics and the literature reviewed. This study highlights potential harmful or ineffective effects of this combination since no adequate evidence exists about its off-label use.


Introduction
The estimated prevalence of chronic pain in the US is 11% [1] and 19% in Europe [2]. Not only the US [3] but also some European countries such as Germany [4] and United Kingdom [5] have shown a signi cant increase in the consumption of opioid analgesics and the trends in mortality due to opioid poisoning pointing out similarities to the US. So far, the key points that have raised are the increased risk of misuse (diversion), overdose, death, abuse, dependence, bone fractures, myocardial infarctions, constipation and sexual dysfunction [6].
A recent study about the prescribing patterns of opioids from 2013 to 2017 in one of the main health administrative regions in Portugal showed an increase approximately 1.5-fold over a 4-year period (2013-2017) [7]. In Spain, the use of opioids for the period 2008-2015 [8] went from 7.3 de ned daily doses (DDD) per 1,000 inhabitants per day to 13.3. The appearance on the market of tramadol/dexketoprofen 75 mg/25 mg lm-coated tablets in 2016 and the increase of consumption in 2017 are noteworthy [9]. According to its summary of product characteristics (SmPC) [10] the authorised therapeutic indication is symptomatic short-term treatment of moderate to severe acute pain in adult patients whose pain is considered to require a combination of tramadol and dexketoprofen. The recommended dosage is one tablet with a minimum dosing interval of 8 hours, without exceeding three tablets per day and ve days of treatment. Switching to a single agent analgesia should be considered according to pain intensity and response of the patient.
Despite the progressive change and the increasing use observed that can impact of pain as a public health issue, there have not been published studies on the use of the combination tramadol/dexketoprofen in clinical practice.

Aim
We aimed to describe the pattern of use of xed-dose combination of tramadol/dexketoprofen in the eld of primary health care to examine potential off-label prescribing and warn about possible risks and suggest interventions.

Ethics Approval
Ethical approval was not required as it was made a secondary analysis in accordance with Spanish regulations. Total population were patients covered by the DAP Costa de Ponent with an active prescription of tramadol/dexketoprofen on March 28, 2018 (cut-off date chosen as a cross sectional data to provide a snapshot of drug use at a day). Target population were those patients who were prescribed tramadol/dexketoprofen over 20 day's treatment, since it was considered 20 days as the limit for a shortterm treatment of acute pain.

Methods
The computerized health records of patients with active tramadol/dexketoprofen prescription were selected from the e-CAP computer program, and extracted from anonymized data.
The variables studied were demographic (age, sex) and clinical (prescribers' specialty; characteristics of the study drug (combined tramadol/dexketoprofen) regarding indications, dosing and duration of treatment; diagnoses of target patients; previous analgesic treatment and concomitant analgesic treatment).
A descriptive statistical analysis of the data was carried out. Discrete variables were published as proportions or frequencies and continuous variables as means and standard deviations.

Results
On the cut-off date, there were 176 patients with active prescription of tramadol/dexketoprofen. All these patients (100%) had a treatment duration exceeding 5 days and 72.7% (N = 128) exceeding 20 days; 73.4% of them (N = 94) were women. Figure 1 (a-d) shows the distribution of patients according to the origin of prescriptions, dosing, duration of treatment and diagnoses.
Twelve patients did not have any diagnosis speci ed at e-CAP. Table 1 shows in detail the diagnoses classi ed as "Other" in Fig. 1d. Only three patients out of 176 (1.7%) had a treatment associated with an oncological diagnosis. Concerning medication prior to tramadol/dexketoprofen associated with the treatment of pain, 14.1% of patients (N = 18) had not taken any medication for pain before. In contrast, 18.8% (N = 24) had been previously treated with one medication for pain, 25.8% (N = 33) had been previously treated with 2 drugs and 41.4% (N = 53) had been previously treated with more than two drugs. Finally, 7.8% of patients (N = 10) had been previously treated with tramadol and dexketoprofen separately.
Regarding the associated pain medication prescribed concurrently with tramadol/dexketoprofen, 14.1% of patients (N = 18) had not ever taken any medicine for pain at any time during treatment with tramadol/dexketoprofen. 26.6% of patients (N = 34) had been treated with one medicine for pain in addition to tramadol/dexketoprofen, 24.2% (N = 31) with two and 35.1% of patients (N = 45) with more than two. Table 2 shows the active substances and/or therapeutic groups that had been prescribed concurrently with tramadol/dexketoprofen.

Discussion
Current evidence indicates that opioids should not be used as rst line therapy, but only when all other treatment options have been tried and failed [12]. The National Institute for Health and Care Excellence's (NICE) found no evidence of long-term effectiveness in chronic non-cancer pain treatment, despite its analgesic effect in the short-term. [13] Furthermore, prescribing of opioids for acute indications such as surgical pain in opioid-naïve patients can also lead to a transition to chronic opioid therapy. [5] A higher percentage of women with active prescription and an average age of 54.7 years were expected data since the prevalence of pain for more than three months is more common in women and in people over 65 years old in Spain.
Exploring the origin of the prescriptions, most of them came from the general practitioner, when a prescription by a specialist would be advisable [14]. The in uence of general practitioner on the dissemination of opioids is substantial to promote an appropriated use. Insu cient training and information about opioid management protocols, time pressure to assess patients properly and systemlevel constraints such as limited access to specialized and integrative care are some of the reasons that could explain the off-label prescription of opioids [14].
About the duration of treatment, 36.7% of patients had a prescription with durations between 9.5 and 12 months. According to the SmPC of tramadol/dexketoprofen [10], it is intended only for short-term use and treatment should be limited strictly to the symptomatic period (5 days). Although in our study, only three patients had post-surgical pain treatment and for 10 days, several studies of chronic postsurgical pain [14] showed that if postoperative pain is not managed properly, it could evolve in chronic pain, becoming a serious problem. In addition, in the absence of evidence on the e cacy and safety of long-term opioid use, it is recommended to re-evaluate at three months the opioid treatment [12].
In relation to the diagnoses found, tramadol/dexketoprofen was hardly used in cancer pain management or in the context of palliative care what is a cause of concern considering the long-term treatment observed (more than 300 days) in 43% of patients. In reference to the most common diagnoses (primary osteoarthritis, low back pain, bromyalgia and lumbago with sciatica), nociceptive pain was highly prevalent and should be treated with analgesics and anti-in ammatory drugs. Although long-term opioid treatment has not yet been widely studied on a population basis, many patients may tolerate and respond to this treatment, and it should not be denied to them, with a close follow-up in terms of effectiveness and safety. [15] It was also found that 26.6% of patients with tramadol/dexketoprofen prescription had been treated concomitantly with one of the two components of this drug, which was unable to account for.
Interventions to improve prescribing should be carried out since there are numerous contraindications and precautions related to tramadol/dexketoprofen, as well as risk of tolerance and addiction [10]. In this sense, some interventions could be that patients had an assessment of their pain and their perspectives regarding the use of this drug (and opioids in general), and physician considered them to promote their correct use, specially preventing their misuse [12]. Another possibility could be to make a reminder to physicians about the indications of tramadol/dexketoprofen and provide information about the inadequate diagnoses and/or prolonged use in chronic noncancer pain, and alternatives in such cases according to evidence [5].

Conclusion
The present study points out that tramadol/dexketoprofen is frequently used off-label, as treatments were usually prescribed without a diagnosis associated with acute pain and were lengthened beyond the days authorised and in most cases 4 times longer than the time period indicated. These results warn about potential harmful or ineffective effects ("concern") of tramadol/dexketoprofen combination since no adequate evidence exists about its off-label use. Interventions to improve tramadol/dexketoprofen prescribing for the treatment of moderate-to-severe acute pain are needed.