Servicios personalizados
Inglés (pdf)
Articulo en XML
Referencias del artículo
Traducción automática
Enviar articulo por email
Citado por SciELO 
Citado por Google 
Similares en
SciELO 
Similares en Google 
Permalink
Background
Nonsteroidal anti-inflammatory drugs (NSAIDs) are a heterogeneous group of drugs that are among the most widely used due to their anti-inflammatory, analgesic and antipyretic action1. The most commonly used drug is ibuprofen, followed by naproxen2. Patients with chronic pain are one of the most frequent users of NSAIDs. The World Health Organization (WHO) introduced the analgesic ladder for cancer pain3, which later extended to chronic non-malignant pain. In this scale, NSAIDs are described as the main option for the management of pain in the first step of the ladder, and as a complementary medication in the second and third steps, adding weak or strong opioids and adjuvants4.
It must be taken into account that NSAIDs are considered inappropriate in 25% of patients, and in 50% in patients aged over 65; the probability that the prescription is inappropriate is five times higher in this age group5. Several clinical practice guidelines warn about prescribing NSAIDs to patients with cardiovascular disease, chronic kidney disease, hypertension, heart failure and cirrhosis; and recommend caution if the prescription of NSAIDs is necessary1,6. The authors have not found in the scientific literature a document that synthesizes the evidence on the different aspects of the safety of NSAIDs.
For all that, this study aims at providing a wide synthesis of the available evidence regarding the safety of NSAIDs in patients with cardiovascular disease, chronic kidney disease, hypertension, heart failure, or liver cirrhosis.
Methods
An umbrella review (review of existing systematic reviews and meta-analyses) was carried out7. The PICO model was followed for this review:
- P (population): patients with cardiovascular comorbidities or at risk of suffering complications, and general population
- I (intervention): use of NSAIDs
- C (comparison): any
- O (outcome): safety.
The databases consulted were Medline, Embase, Scopus and the Cochrane Library. The search was complemented with the following resources: Google Scholar and the Spanish Network of Health Technology Assessment Agencies8. In addition, the reference lists of the identified articles were examined. The search strategy adopted for the different databases was validated by a librarian specialized in Public Health (Appendix 1) and was updated to 1st July, 2019. To maximize search sensitivity, two searches were carried out in parallel, with and without review filter (SIGN). The literature search was complemented by other databases and grey literature. The writing process of this umbrella review was done following an adaptation of the guide published by Bougioukas et al.9.
Inclusion criteria: systematic reviews (SR) and meta-analyses offering relevant information concerning the object of the study, reviews written in English or Spanish, with no publication date restrictions; when two reviews with similar objectives were located, the most recently published one was selected. Exclusion criteria: original article, editorial or commentary, animal/laboratory study, or NSAID exposure not clear.
References were stored in Mendeley reference manager and reviewed to identify potentially relevant documents. Two reviewers independently screened titles (AOL and ESF) and abstracts to determine whether the documents met the inclusion criteria; those that met the criteria were kept as potential documents (first screening step) (Appendix 2). Additional references were obtained after screening by cross-checking the references of previously identified documents. Both reviewers then independently evaluated the papers selected for full-text examination (second screening step). A third independent reviewer (CBT) resolved any discrepancies during both screening processes. The analysis of the content was performed according to categories related to the different predefined populations. Finally, a narrative synthesis was done.
A measurement tool to assess systematic reviews (AMSTAR)10 was chosen to assess the quality of the reviews included in our umbrella review. AMSTAR was developed to evaluate SR of randomized trials and allows a more detailed evaluation of SR that also include non-randomized studies of health interventions, which are increasingly incorporated into SR.
Results
The search identified a total of 1,406 references (Figure 1). After a first reading of titles and abstracts, a total of 103 reviews were selected to the next step for full-text screening to ensure articles met all eligibility criteria. Finally, 22 articles were included in this overview (Table 1). The authors reviewed the possibility of performing a meta-analysis with the results of the articles included in this review; however, it was not possible due to the differences in the populations of the included reviews.
Table 1. Characteristics of the systematic reviews and meta-analysis included in this umbrella review
| Authors | Studies included in the review |
|---|---|
| Sriuttha et al.11 | 18 RCT (45,705 participants) |
| Nderitu et al.12 | 7 studies: 5 CO, 1 CC, 1 CS |
| Zhang et al.13 | 15 studies: |
| - 10 studies with general population (n = 1,609,136): 5 nCC, 5 CC | |
| - 5 studies with chronic renal failure patients (n = 106,681), CC | |
| Asghar et al.14 | 19 studies: 12 CC, 4 CO, 3 RCT |
| Yaxley et al.15, | 9 studies: 5 CO, 4 CC (12,418 participants and 23,877 controls) |
| Villa et al.16 | 19 studies |
| Ungprasert et al.17 | 7 studies: 4 CC, 3 CO (7,543,805 participants) |
| Liu et al.18 | 5 studies: 3 CC, 2 CO |
| Seshasai et al.19 | 9 RCT (over 100,000 participants) |
| Whitlock et al.20 | 7 studies: 6 CO, 1 RCT |
| Boulakh et al.21 | 14 studies: 2 dbRT, 10 CO, 2 CC |
| Ungprasert et al.22 | 6 studies: 3 CC, 3 CO |
| García Rodríguez et al.23 | 4 CO, 2 nCC |
| Salvo et al.24 | 29 MA |
| Scott et al.25 | 21 studies: 5 CC, 2 CO, 12 RCT |
| Varas-Lorenzo et al.26 | 25 studies: 8 CO, 14 nCC, 3 CC |
| McGettigan et al.27 | 51 studies: |
| - 30 CC (184,946 cardiovascular events) | |
| - 21 CO (2.7×106 exposed individuals). | |
| Mackenzie et al.28 | 17 RCT |
| Major et al.29 | 3 clinical trials (4,468 participants and 16,740 person-years of follow-up) |
| Luni et al.30 | 6 studies: 1 CO, 5 CC |
| Lee et al.31 | 4 RCT |
| Bem et al.32 | 28 controlled studies |
RCT: randomized clinical trial; CO: cohort; CC: case-control; CS: cross-sectional; nCC: nested case-control in a cohort; dbRT: randomized double-blind trial; MA: meta-analysis.
Patients with risk of hepatotoxicity
Sriuttha et al.11 included 18 randomised controlled trials (RCT) (n = 45,705 subjects) that evaluated the risk of hepatotoxicity associated with NSAIDs. Length of the studies ranged from four weeks or less for four studies (22.2%), six weeks for one study (5.6%), 12 weeks for six studies (33.3%), and 16 weeks or longer for seven studies (38.9%). Indication of NSAIDs corresponded to osteoarthritis (13 studies) and for rheumatoid arthritis (five studies). A total of eight studies describing the use of three NSAIDs (celecoxib, etoricoxib and diclofenac) reported clinically significant hepatotoxicity, with diclofenac showing the highest proportion of hepatotoxic events (range 0.015 × 10-2 - 4.3 × 10-2, followed by celecoxib (range 0.13 × 10-2 - 0.38 × 10-2), and etoricoxib (range 0.005 × 10-2 - 0.930 × 10-2).
Patients with risk of chronic kidney disease (CKD)
Nderitu et al. review12 included seven studies, of which three were included in the meta-analysis. This study found that the use of NSAIDs in regular doses did not significantly affect the risk of accelerated progression of CKD (OR = 0.96; 95%CI: 0.86-1.07), but the use of high doses of NSAIDs increased the risk of accelerating the progression of CKD (OR = 1.26; 95%CI: 1.06-1.50).
The second review13 examined the risk of acute kidney injury (AKI) - defined by: change in estimated glomerular filtration rate (eGFR) or creatinine clearance (CrCl) or serum creatinine (SCr)) associated with NSAIDs in the general population and in those with pre-existing CKD. This review included 10 articles targeting the general population (1,609,136 participants), of which were performed with patients with chronic renal failure (n = 106,681). Regarding the design, five were case-control studies and the rest were nested case-control studies. Eight of the 10 studies targeting the general population showed a significant association between exposure to NSAIDs and kidney damage. The meta-analysis showed an OR = 1.73 (95%CI: 1.44-2.07), although a high heterogeneity was found (I2 = 89%, p <0.001). In the case of patients with kidney disease, four of the five studies showed a significant association (OR = 1.63; 95%CI: 1.22-2.19; I2 = 71%). In the subgroup analysis, it was found that a greater cyclooxygenase-2 (COX-2) selectivity was associated with a lower risk of AKI. Similar results were shown in the review of Asghar et al.14, which demonstrated that meloxicam did not elevate renal risk (OR = 0.99; 95%CI: 0.72-1.35) in general population.
In order to assess the relation between NSAIDs and analgesic nephropathy, a total of nine studies were included (12,418 participants and 23,877 controls) in Yaxley et al. review15, and only one of the nine included studies found a relation between the chronic use of NSAID and analgesic nephropathy (OR = 2.1; 95%CI: 1.1-4.1).
Patients on antihypertensive treatment
A review to assess the interaction between antihypertensives and NSAIDs (n = 19 studies) found 21 pharmacodynamic interactions, which according to their clinical relevance were classified in level-2 high risk (76.2%) and level-3 medium risk (23.8%). Results showed that some NSAIDs might attenuate the effectiveness of antihypertensive treatment, especially with angiotensin-converting enzyme inhibitors, diuretics, beta-blockers and angiotensin receptor blockers. There was no evidence of effect modification of calcium channel antagonists, especially dihydropyridine, by concurrent use of NSAIDs16.
Patients with cardiovascular disease (CVD) - Primary prevention
Ungprasert et al.17 reviewed the association between the use of NSAIDs and the incidence of heart failure (n = 7 studies with 7,543,805 participants). A significantly higher risk of developing heart failure was found with the use of NSAIDs (RR = 1.17; 95%CI: 1.01-1.36; I2 = 53%). When performing the subgroup analysis, a higher risk of heart failure was found among the users of conventional NSAIDs (RR = 1.35; 95%CI: 1.15-1.57; I2 = 0%), but not among COX-2 specific inhibitors (COXIB) users (RR = 1.03; 95%CI: 0.92-1.16; I2 = 0%). The use of NSAIDs18, excluding aspirin, was associated with a 12% increased risk for atrial fibrillation (RR = 1.12; 95%CI: 1.06-1.18; I2 = 65%) in five studies (three case-controls and two cohorts). Selective NSAIDs showed an increased risk of atrial fibrillation (RR = 1.24; 95%CI: 1.18-1.30).
Another systematic review19, involving over 100,000 participants and with a mean follow-up of 6.0 years, concluded that aspirin treatment reduced total CVD events (OR = 0.90; 95%CI: 0.85-0.96; NNT = 120), driven primarily by reductions in non-fatal myocardial infarction (OR = 0.80; 95%CI: 0.67-0.96; NNT = 162), and no significant reduction was found in CVD death (OR = 0.99; 95%CI: 0.85-1.15). In secondary prevention, this review found that the aspirin in prophylactic treatment does not lead to reductions in cardiovascular death.
With regard to the use of aspirin, another SR concluded that an intervention of very low-dose aspirin (≤100 mg daily or every other day) could increase haemorrhagic stroke risk although results are not statistically significant (OR = 1.27; 95%CI: 0.96-1.68)20.
Patients with CVD - Secondary prevention
Regarding population with myocardial infarction, Boulakh et al.21 included a total of 14 articles, with a follow-up between 30 days and 15 years. This review showed a clear relationship between the use of NSAIDs and re-infarction and death in patients with previous myocardial infarction. Ten of the included studies analysed re-infarction and death and provided evidence of a clear relationship between the use of NSAIDs and re-infarction and death in patients with previous myocardial infarction, being the risk higher with celecoxib, ibuprofen and diclofenac than with naproxen. Naproxen showed the lowest risk of cardiovascular death: in one study the HR was 1.27 (95%CI: 1.01-1.59) and in other they found no association to death or recurrent infarction when the patients were treated with naproxen and ibuprofen.
In relation to heart failure, the review of Ungprasert et al.22 found that in patients with heart failure, the use of conventional NSAIDs was associated with an increased risk of exacerbation of heart failure (RR = 1.39; 95%CI: 1.20-1.62). In concordance with these findings, Liu et al.18) found an increased risk for atrial fibrillation (RR = 1.82; 95%CI: 1.42-2.32) in patients under treatment with NSAIDs. Patients with CKD also showed an increased risk for atrial fibrillation (RR = 1.58; 95%CI: 1.34-1.85).
Patients with CVD - General population
SR by García-Rodríguez et al.23 included six studies with different proportions of patients in secondary prevention. The review concluded that the overall effect of NSAIDs for fatal and non-fatal myocardial infarction was RR = 1.21 (95%CI: 1.07-1.37), being the risk for non-fatal myocardial infarction higher than the risk for fatal myocardial infarction (RR = 1.30; 95%CI: 1.20-1.41 vs. RR = 1.02; 95%CI: 0.89-1.17).
To estimate incidence rates for cardiovascular adverse events associated with the use of NSAIDs, a total of 29 meta-analyses were analysed by Salvo et al.24, providing 109 incidence rates. Cumulative incidence rates for myocardial infarction ranged from 0.12-1.35% for celecoxib, 0.24-0.66% for etoricoxib, 0.09-0.37% for lumiracoxib, 0.30-0.89% for rofecoxib, and 0.11-0.61% for valdecoxib; there was a single estimate for parecoxib/valdecoxib combination (0.58%). Moreover, cumulative incidence rates for stroke ranged from 0.02 to 0.79% for celecoxib and from 0.06 to 0.34% for lumiracoxib; there was a single estimate for rofecoxib (0.12%) and for valdecoxib (0.18%).
Scott et al.25 concluded that NSAIDs increased the risk of heart failure by 30-100%. The absolute risk is small; less than one patient developed heart failure attributable to NSAIDs per hundred patient-years of treatment. These risks were relatively small and similar with conventional NSAIDs and COXIBs (OR = 1.14; 95%CI: 0.85-1.53). The risk of heart failure in patients receiving NSAIDs was higher in those with pre-existing heart failure.
Another SR that assessed the risk of acute myocardial infarction (AMI) associated with the use of individual NSAIDs in 25 studies (100,000 AMI events)26, included low-medium to high risk patients according to the prior history of myocardial infarction or coronary heart disease. There were no statistically significant differences of AMI ranged from naproxen, followed by celecoxib and ibuprofen; while an increased risk was shown for indomethacin (RR = 1.40; 95%CI: 1.21-1.62), etodolac (RR = 1.55; 95%CI: 1.16-2.06), and etoricoxib (RR = 1.97; 95%CI: 1.35-2.89). Furthermore, except for naproxen, a higher risk was associated with higher doses and in people with prior history of coronary heart disease, even for short-time (≤3 months) use of the NSAID.
The review conducted by McGettigan et al.27, including RCTs and observational studies (51 papers), reached similar results.
The review published by Mackenzie et al.28 suggested that there were no significant differences in cardiovascular outcomes between lumiracoxib and placebo or between lumiracoxib and other NSAIDs in patients with osteoarthritis. Accordingly, Asghar et al.14 (with 19 included studies: 12 case-controls, four cohorts and three RCTs) showed that meloxicam was associated with a low increase in composite of vascular and myocardial risk (OR = 1.14; 95%CI: 1.04-1.25), which was mainly vascular in nature (OR = 1.35; 95%CI: 1.18-1.55) as it did not elevate the risk of myocardial infarction (OR = 1.13; 95%CI: 0.98-1.32).
Major et al.29 included three trials with 4,468 participants and 16,740 person-years of follow-up. The review concluded that aspirin did not achieve significant reductions in the risk of major cardiovascular events (RR = 0.92; 95%CI: 0.49-1.73, I2 = 71%) or mortality (RR = 0.74; 95%CI: 0.55-1.00; I2 = 0%) compared to the control group which was receiving placebo. Another SR including six observational studies30 concluded that there was no difference in major adverse cardiovascular events with planned discontinuation of aspirin (OR = 1.17; 95%CI: 0.76-1.81; I2 = 55%). A SR31 that included RCTs comparing warfarin and aspirin, carried out in patients with heart failure and normal sinus rhythm, concluded that there were no significant differences in the death-stroke endpoint (RR = 0.94; 95%CI: 0.84-1.06; p = 0.31), or death by any cause (RR = 1.01; 95%CI: 0.89-1.14; p = 0.89) between warfarin and aspirin. On the contrary, the use of warfarin was associated with a lower risk of any stroke (RR = 0.56; 95%CI: 0.38-0.82, p = 0.003) and ischemic stroke (RR = 0.45; 95%CI: 0.24-0.86, p = 0.02) compared to aspirin.
Finally, a review32 assessed the evidence on the effects of different aspirin regimens in terms of timing (chronotherapy) or frequency of dosing in the prevention of cardiovascular disease. In general, the lack of significant differences was highlighted in both primary and secondary prevention to evaluate the evidence.
Risk of bias in the studies included
The evaluation of methodological quality of the studies included in this review by the AMSTAR scale yielded heterogeneous risk of bias (Table 2). The. The quality of the review was critically low for 12 of the included reviews (54.5%). In the rest of the cases the quality was low. The issues less addressed were To provide a list of excluded studies and Justify the exclusions (critical) and the sources of funding for the studies included in the review.
Table 2. AMSTAR 2 assessment of the included systematic reviews and meta-analysis
| AMSTAR questions* | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Authors | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | Rating |
| Sriuttha et al.11 | √ | P | √ | P | √ | √ | - | P | P | - | n/a | n/a | - | - | n/a | √ | CL |
| Nderitu et al.12 | √ | P | √ | P | √ | √ | - | P | P | - | √ | - | √ | √ | √ | √ | L |
| Zhang et al.13 | √ | P | √ | P | √ | √ | - | P | - | - | √ | √ | √ | √ | √ | √ | L |
| Asghar et al.14 | √ | P | √ | P | √ | √ | - | P | - | - | n/a | n/a | - | √ | n/a | √ | CL |
| Yaxley et al.15 | √ | P | √ | P | √ | √ | - | P | - | - | n/a | n/a | - | - | n/a | √ | CL |
| Villa et al.16 | √ | P | √ | P | √ | √ | - | P | - | - | n/a | n/a | - | - | n/a | √ | CL |
| Ungprasert et al.17 | √ | P | √ | P | √ | √ | - | P | √ | - | √ | √ | √ | √ | √ | √ | CL |
| Liu et al.18 | √ | P | √ | P | √ | √ | - | P | √ | - | √ | √ | √ | √ | - | √ | L |
| Seshasai et al.19 | √ | P | √ | P | √ | √ | - | P | √ | - | √ | √ | √ | √ | √ | √ | L |
| Whitlock et al.20 | √ | P | √ | P | √ | √ | - | P | - | - | n/a | n/a | - | - | n/a | √ | CL |
| Boulakh et al.21 | √ | P | √ | P | √ | √ | - | P | - | - | n/a | n/a | - | - | n/a | √ | CL |
| Ungprasert et al.22 | √ | P | √ | P | √ | √ | - | P | √ | - | √ | √ | √ | √ | √ | √ | L |
| García Rodríguez et al.23 | √ | P | √ | P | √ | √ | - | P | - | - | n/a | n/a | - | √ | n/a | √ | CL |
| Salvo et al.24 | √ | P | √ | P | √ | √ | - | P | √ | - | n/a | n/a | √ | - | - | √ | L |
| Scott et al.25 | √ | P | √ | P | √ | √ | - | P | - | √ | √ | - | - | - | - | √ | CL |
| Varas-Lorenzo et al.26 | √ | P | √ | P | √ | √ | - | P | √ | - | √ | √ | √ | √ | √ | √ | L |
| McGettigan et al.27 | √ | P | √ | P | √ | √ | - | P | √ | - | n/a | n/a | √ | √ | n/a | √ | L |
| Mackenzie et al.28 | √ | P | √ | P | √ | √ | - | P | √ | - | √ | √ | √ | - | √ | √ | L |
| Major et al.29 | √ | P | √ | P | √ | √ | - | P | √ | - | √ | √ | √ | √ | √ | √ | L |
| Luni et al.30 | √ | P | √ | P | √ | √ | - | P | √ | - | √ | √ | √ | √ | √ | √ | L |
| Lee et al.31 | √ | P | √ | P | √ | √ | - | P | √ | - | √ | √ | - | √ | - | √ | CL |
| Bem et al.32 | √ | P | √ | P | √ | √ | - | P | √ | - | n/a | n/a | √ | n/a | √ | √ | L |
*: 1. Did the research questions and inclusion criteria for the review include the components of PICO? 2. Did the report of the review contain an explicit statement that the review methods were established prior to the conduct of the review and did the report justify any significant deviations from the protocol? 3. Did the review authors explain their selection of the study designs for inclusion in the review? 4. Did the review authors use a comprehensive literature search strategy? 5. Did the review authors perform study selection in duplicate? 6. Did the review authors perform data extraction in duplicate? 7. Did the review authors provide a list of excluded studies and justify the exclusions? 8. Did the review authors describe the included studies in adequate detail? 9. Did the review authors use a satisfactory technique for assessing the risk of bias (RoB) in individual studies that were included in the review? 10. Did the review authors report on the sources of funding for the studies included in the review? 11. If meta-analysis was performed did the review authors use appropriate methods for statistical combination of results? 12. If meta-analysis was performed, did the review authors assess the potential impact of RoB in individual studies on the results of the meta-analysis or other evidence synthesis? 13. Did the review authors account for RoB in individual studies when interpreting/ discussing the results of the review? 14. Did the review authors provide a satisfactory explanation for, and discussion of, any heterogeneity observed in the results of the review? 15. If they performed quantitative synthesis did the review authors carry out an adequate investigation of publication bias (small study bias) and discuss its likely impact on the results of the review? 16. Did the review authors report any potential sources of conflict of interest, including any funding they received for conducting the review? Rating overall confidence: H (high): zero or one non-critical weakness; M (moderate): more than one non-critical weakness; L (low): one critical flaw with or without non-critical weaknesses; CL (critically low): more than one critical flaw with or without non-critical weaknesses. √: yes; P: partial yes; -: no; n/a: not applicable because the study is not a meta-analysis.
Discussion
This umbrella review can be a useful tool for offering evidence to physicians in the prescription process considering the different aspects of safety and efficacy of NSAIDs33.
IT compiling the evidence from 22 existing reviews and meta-analyses highlights the need for caution when prescribing NSAIDs to patients with prior cardiovascular disease, hypertension, at risk of cirrhosis or kidney disease, and general population. Our review found that the use of NSAIDs was associated with an increased likelihood of cardiovascular events17,18,22,23,25,26,34 and cardiovascular risk27. The reviewed evidence showed that NSAIDs were associated with cardiovascular side-effects that increase both systolic and diastolic blood pressure and this may precipitate cardiovascular events35-38. It is important to underline that the simultaneous use of NSAIDs and antihypertensives can increase blood pressure levels, generating problems of ineffectiveness of antihypertensive treatments in some patients16.
On the other hand, a key element that is generating a great debate in the scientific community is the use of aspirin for primary prevention of cardiovascular disease in elderly patients. We have found two SR analysing this issue19,20, with varying results. Seshasai et al.19 concluded that aspirin treatment reduced total CVD events, driven by reductions in non-fatal myocardial infarction, but no significant reduction was found in CVD death, while Whitlock et al.20 found that an intervention of very low-dose aspirin (≤100 mg daily or every other day) could increase haemorrhagic stroke risk although results were not statistically significant.
Although the association between NSAIDs and kidney damage was not clearly established, except for NSAID at high doses, this potential relationship becomes highly relevant if we take into account the wide use of NSAIDs among our community and the fact that 2% of the national health budget is spent on renal replacement therapy39.
In general, this overview could be affected by the lack of representability. Many of the studies included in this analysis showed increased risks in general populations by assessment of harms in relative terms (RR or OR), with small absolute differences when they were quantified. Data for risk increment in absolute terms in patients previously affected by renal, hepatic or cardiovascular conditions are lacking in some reviewed studies. In these settings with augmented basal risk, it would be reasonable to assume greater absolute risks if relative increases were at least maintained.40
Finally, it is believed that the large number of SR included in this umbrella review provided a good estimation of the state of the art of recommendations for prescribing NSAIDs41. Moreover, taking into account the results of the analyses, and that the risks are usually minimised, a cautious attitude must be taken.
Our findings are in line with the guidelines recommending avoidance of NSAIDs in patients with cardiovascular disease, chronic kidney disease, hypertension and cirrhosis. Taking into account the different risk profiles of patients and the differences that exist between the drugs, it is necessary to individualize their use based on the type of process for which they are used and the characteristics of the patient42. Caution should therefore be taken when recommending or prescribing NSAIDs and patients should be advised about its potential consequences on health. Proper follow-up should be ensured.
The conclusion of this review of reviews is that the prescription of NSAIDs should continue to be done with caution, taking into account the risk of NSAIDs in high-risk patients. This conclusion is in concordance with other international scientific initiatives such as Choosing wisely in the United States43, Do not do44 from The National Institute for Health and Care Excellence (NICE) in the United Kingdom and Commitment to the Quality of Scientific Societies45 in Spain.
