DOI: https://doi.org/10.21203/rs.3.rs-250344/v1
To assess the efficacy and safety of prasugrel at low doses compared to clopidogrel by looking at the occurrence of major adverse cardiac events (MACE) and major bleeding in patients with acute coronary syndrome (ACS) or undergoing percutaneous coronary intervention (PCI).
We searched PubMed, EMBASE, the Cochrane Central Register of Controlled Trials, and ClinicalTrials.gov for eligible randomized controlled trials (RCTs) and observational studies assessing efficacy and safety of low-dose prasugrel versus clopidogrel in patients with ACS or undergoing PCI up to May 22, 2020. We did a meta-analysis using a random-effects model to estimate relative risks (RRs). The primary efficacy and safety endpoints were MACE and major bleeding, respectively.
Six RCTs (n = 6,131) and six observational studies (n = 31,426) were included. There was no MACE reduction in patients receiving low-dose prasugrel compared with those receiving clopidogrel (RR 1.02, 95%CI 0.91 to 1.14), but there was an increased risk of major bleeding (RR 1.35, 95%CI 1.10 to 1.67).
Low-dose prasugrel yields no increase in efficacy when compared with clopidogrel, but it does expose patients to an increased risk of bleeding. Most studies considered here were conducted in Japan. Studies conducted with non-Asian patients may find that low-dose prasugrel offers a more favorable efficacy and risk profile. Considering the results of this analysis we believe low-dose prasugrel should be prescribed with extreme caution as it may result in bleeding events without any additional benefit over clopidogrel.
Dual antiplatelet therapy (DAPT) combining aspirin and a P2Y12 inhibitor has become an essential part of acute coronary syndrome (ACS) treatment, particularly for patients undergoing percutaneous coronary intervention (PCI) [1, 2]. The use of DAPT reduces the risk of major ischemic events. [2, 3]. Clopidogrel has been a widely used P2Y12 inhibitor used in DAPT, but slow onset of action, variability in patient responses, and problems with drug resistance have led to pressure to develop new P2Y12 inhibitors [4, 5].
Prasugrel is a third-generation thienopyridine antiplatelet drug that provides more rapid onset and greater inhibition of platelet function than clopidogrel [6]. Because of these advantages, the new ESC clinical practice guidelines for the management of ACS in patients presenting without persistent ST elevation recommend prasugrel as a preferred P2Y12 inhibitors in ACS patients undergoing PCI [3]. The standard dose of prasugrel has been shown to reduce the risk of ischemic events, including stent thrombosis compared with clopidogrel in patients with ACS undergoing PCI [7]. However, in some instances prasugrel may be too effective at inhibiting platelet function, and patients receiving the drug have exhibited severe and even fatal bleeding. Indeed, no net clinical benefit was observed in elderly patients (≥ 75 years of age), patients with low body weight < 60 kg [7], or in patients of Asian origin due to the increased risk of bleeding in these populations [8, 9]. Hence, protocols for the use of prasugrel have rapidly adopted lower recommended doses for patients with ACS, particularly in East Asian populations to mitigate bleeding risk.
Previous randomized controlled trials (RCTs) have demonstrated that low-dose prasugrel has a similar efficacy and safety profile to clopidogrel. In the TRILOGY-ACS study, reduced-dose prasugrel (loading dose of 60 mg and maintenance dose of 5 mg) was associated with a similar risk of ischemic events and bleeding events compared with clopidogrel in an elderly population with ACS [10]. In the PRASFIT-ACS study, an even lower dose of prasugrel (loading dose of 20 mg and maintenance dose of 3.75 mg) provided similar efficacy without an increased risk of bleeding compared with clopidogrel, suggesting that risks and benefits of the drug can be effectively balanced in an East Asian population [11]. However, large observational studies in Japan found that low-dose prasugrel was associated with a significantly increased risk of short-term bleeding, and patients in these studies did not exhibit any beneficial reduction in ischemic events. In fact, patients treated with low-dose prasugrel trended towards having more ischemic events than those treated with clopidogrel [12, 13]. It is not clear why these studies have found conflicting results, and it remains a major issue to understand the relationship between the risks and benefits of prasugrel at low doses.
Honing protocols used for prasugrel administration is now a pressing issue as this drug is likely to be used widely because of its status as a preferred P2Y12 inhibitors [3]. To address this issue, we performed a systematic review and meta-analysis to assess the risks and benefits of prasugrel at low doses compared to clopidogrel by looking at the occurrence of major bleeding and major adverse cardiac events (MACE) in patients with ACS or undergoing PCI.
This study was done in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines for meta-analyses of interventional studies [14], and observational studies [15]. The study protocol was registered on PROSPERO (CRD42020188261) [16].
Our systematic review focused on patients with ACS or patients who undergoing PCI. The intervention of interest was low-dose prasugrel (60/5 mg loading/maintenance dose or 20/3.75 mg loading/maintenance dose) for long-term therapy.
We search PubMed, EMBASE, the Cochrane Central Register of Controlled Trials, and ClinicalTrial.gov from inception until May 22, 2020. Our search terms were: (“myocardial infarction” OR “coronary artery disease” OR “coronary heart disease” OR “ischemic heart disease”) AND prasugrel AND (“cardiac event” OR cardiovascular event* OR cerebrovascular event* OR stent thrombosis OR mortality OR death OR bleeding). We applied no language, publication date, and publication status restrictions. We also hand-searched the bibliographies of included studies and relevant systematic review articles for additional articles. The inclusion criteria were: (1) RCTs or observational studies enrolling patients with ACS or patients who undergoing PCI; (2) comparison of low-dose prasugrel versus standard-dose clopidogrel (300–600/75 mg, loading/maintenance dose); (3) reporting of MACE, death, cardiovascular death, stroke, myocardial infarction (MI), and major or minor bleeding during follow-up. Studies were excluded for any of the following: (1) abstract without full-text or conference proceeding; (2) studies with insufficient data to calculate or extract effect estimates; (3) studies only reporting pharmacodynamic parameters without considering the cardiac endpoints or bleeding. Further details of search strategies are provided in Table S1.
Records were independently screened by title and abstracts to identify potential studies according to the inclusion/exclusion criteria by three investigators (YW, SU or WK). Disagreements were resolved by discussion or a fourth reviewer (KK). Each potentially relevant study was reviewed in full by the same investigators. Data were extracted by two investigators (any two of YW, SU or WK) using standard data extraction forms. The number of events in intention to treat analysis was collected. Adjusted effect estimates (hazard ratio [HR], risk ratio [RR], or odds ratio [OR]) with 95% confidence interval were extracted from observational studies. Any disagreement was resolved by discussion with another reviewer (SS).
The primary efficacy outcome was MACE, defined as the composite of death, non-fatal MI, or non-fatal stroke. The primary safety outcome was major bleeding based on Thrombolysis in Myocardial Infarction (TIMI) criteria [17] or Bleeding Academic Research Consortium (BARC) criteria type 3 or 5 [18]. The secondary endpoint was minor bleeding based on TIMI criteria; and type 1 or 2 bleeding based on BARC criteria [17, 18]. We also evaluated individual components of MACE outcome, and stent thrombosis as additional outcomes. The precise definition of cardiovascular death, non-fatal MI, stent thrombosis, and stroke were determined separately by each study. No substantive differences were apparent in these definitions.
Two investigators (YW, KK) independently appraised the quality of included studies using Cochrane Risk of Bias Tools as appropriate to each study’s design. RCTs were assessed according to the revised Cochrane Risk of Bias Tool for Randomized Trials (RoB version 2.0) [19]. This method considers the study’s randomization process, bias due to deviations from intended interventions, missing outcome data, measurement of the outcome, and selection of the reported result. Each domain was qualitatively classified as at high, medium, or low risk of bias. We used the Risk Of Bias In Non-randomized Studies of Interventions (The ROBINS-I) tool to evaluate the risk of bias in observational studies. This method considers confounding factors, selection of participants, classification of intervention, deviations from intervention, missing data, measurement of outcome and selection of reported results. Each domain was qualitatively classified as at critical, serious, moderate, or low risk of bias. Risk of bias graphs was derived from these tools [20].
We conducted separate analyses for randomized controlled trials and observational studies. We estimated the risk ratios (RRs) with 95% confidence interval (CI) for dichotomous outcomes. Heterogeneity was assessed using the Cochrane Q-statistic and quantified using I2 test with scores below 50% deemed low, 50%-75% deemed moderate and > 75% deemed high heterogeneity [21]. Because we assumed treatment effects varied across trials, we used a random effects model (the DerSimonian and Laird method) to calculate the pooled RRs and 95% Cis [22, 23]. To combine data from observational studies, we obtained adjusted values for log [risk ratio] and log [standard error]. HR and OR were converted to RRs [24]. We assessed small-study effects by using Egger’s test with the finding of a visually asymmetrical funnel plot indicating the existence of bias [25]. We performed subgroup analyses based on clinical characteristics (e.g., dose of prasugrel (3.75 mg or 5 mg), types of clinical events, types of PCI, age groups, and study country). Interactions between subgroups were examined with a test for heterogeneity with alpha set to 0.05. Sensitivity analyses were performed by excluding studies with a high risk of bias to assess the robustness of the findings. Statistical testing was two-sided with alpha set at 0.05.
We used the Grading of Recommendations, Assessment and Evaluation (GRADE) framework to report the overall quality of evidence. The certainty in the evidence for each outcome was graded as high, moderate, low, or very low [26, 27].
Study selection
Overall, 5396 records were identified through database searching. 125 additional records were identified through other sources. After removal of duplicates, a total of 4406 records remained and were initially screened by title and abstract. Of these, 40 were included for full-text review. Twelve studies were included in the final analysis consisting of six RCTs [10, 11, 28–31] and six observational studies [12, 13, 32–35]. The study selection is illustrated in Fig. 1.
Study characteristics
The characteristics of the included studies are summarized in Table 1. Overall, these 12 studies involved 37,557 patients. Six were RCTs (6,131 patients), while the other 6 were observational studies (31,426 patients). The mean age of study participants was 69.9 years (Standard deviation [SD], 5.1) and 24.3% of participants were female (interquartile range [IQR], 19.6% − 33.8%). Eight studies (66.7%) were performed with ACS patients [10–13, 31, 30, 33, 35], and four studies (33.3%) were performed with CAD patients [28, 29, 32, 34]. Nine studies (75.0%) examined the effects of a 3.75 mg dose of prasugrel [10, 12, 13, 28, 31–35], two studies (16.7%) examined 5 mg dose [11, 30] and one (8.3%) compared these two doses [29]. Most studies were conducted in Japan (10 studies) [11–13, 28–30, 32–35], while the other 2 studies included one conducted in Italy [31] and one that gathered data from 52 countries around the world [10]. The median duration of follow-up was 12 months (IQR, 8–12 months). Characteristics of all included studies are given in Table S2.
Study | Study design | Country | Target population | Total population | Patients | Intervention (n) | Control (n) | Outcomes | Follow-up (months) | Age (years), Mean | Female (%) | |
---|---|---|---|---|---|---|---|---|---|---|---|---|
MACE | Major Bleeding | |||||||||||
Isshiki (2014)[28] | RCT | Japan | Asian (Japanese patients) | 742 | CAD (stable angina or prior MI) undergoing PCI | prasugrel 20/3.75 mg (370) | clopidogrel 300/75 mg (372) | Yes | Yes | 12 | 27.6 | 67.5 |
Kimura (2015)[29] | RCT | Japan | Asian (Japanese patients) | 422 | CAD (stable angina or prior MI) undergoing PCI | prasugrel 20/3.75-5 mg (207) | clopidogrel 300/75 mg (104) | Yes | Yes | 3.5 | 14.4 | 64.5 |
Roe (2013)[10] | RCT | Italy | Elderly patients (age ≥ 75 years) | 2083 | ACS (UA/NSTEMI) | prasugrel NA/5 mg (1043) | clopidogrel NA/75 mg (1040) | Yes | Yes | 30 | 50.5 | 79.5 |
Saito (2014)[11] | RCT | Japan | Asian (Japanese patients) | 1363 | ACS (UA/STEMI/NSTEMI) undergoing PCI | prasugrel 20/3.75 mg (685) | clopidogrel 300/75 mg (678) | Yes | Yes | 12 | 21.2 | 65.3 |
Savonitto (2018)[31] | RCT | 52 countries around the world | Elderly patients (age ≥ 75 years) | 1443 | ACS (STEMI/NSTEMI) undergoing PCI | Prasugrel 60/5 mg (713) | clopidogrel 300–600/75 mg (730) | Yes | Yes | 12 | 40.0 | 80.0 |
Kitano (2019)[30] | RCT | Japan | Asian (Japanese patients) | 78 | ACS undergoing PCI | prasugrel 20/3.75 mg (39) | clopidogrel 300/75 mg (39) | Yes | No | 8 | 18.0 | 64.8 |
Akita (2020)[12] | Cohort | Japan | Asian (Japanese patients) | 24032 | ACS undergoing PCI | prasugrel 20/3.75 mg (12016) | clopidogrel 300/75 mg (12016) | No | Yes | In-hospital | 25.9 | 70.6 |
Koyabu (2019)[32] | Cohort | Japan | Asian (Japanese patients) | 500 | CAD (CAD and ACS) undergoing PCI | prasugrel 20/3.75 mg (250) | clopidogrel 300/75 mg (250) | Yes | Yes | 12 | 79.6 | 67.5 |
Ohya (2018)[33] | Cohort | Japan | Asian (Japanese patients) | 992 | ACS undergoing PCI | prasugrel 20/3.75 mg (487) | clopidogrel NA/75 mg (192) | Yes | Yes | 17 | 17.0 | 71.0 |
Shoji (2020)[13] | Cohort | Japan | Asian (Japanese patients) | 1802 | ACS undergoing PCI | prasugrel 20/3.75 mg (901) | clopidogrel 300/75 mg (901) | Yes | Yes | In-hospital | 24.0 | 71.5 |
Tokimasa (2019)[34] | Cohort | Japan | Asian (Japanese patients) | 1031 | CAD undergoing PCI | prasugrel 20/3.75 mg (412) | clopidogrel 300/75 mg (619) | Yes | Yes | 6 | 23.1 | 67.9 |
Yasuda (2019)[35] | Cohort | Japan | Asian (Japanese patients) | 3069 | ACS (STEMI or NSTEMI) undergoing PCI | prasugrel 20/3.75 mg (2607) | clopidogrel 300/75 mg (462) | Yes | Yes | 12 | 23.5 | 69.8 |
Risk of bias
The risk of bias in five RCTs (83.3%) was rated as low risk [10, 11, 28, 29, 31]. We judged the risk of bias as high in one RCT (16.7%) due to inadequate information about allocation concealment in the randomization process [30]. The risk of bias in three observational studies (50.0%) was considered moderate [12, 13, 35], and the other three studies (50.0%) were considered to have serious risk of bias due to the existence of possible confounds (Figure S1) [32–34],
MACE
MACE was reported in six RCTs (n = 6020) [10, 11, 28–31], and five observational studies (n = 7081) [13, 32–35]. There was no difference in the risk of MACE between patients receiving low-dose prasugrel and patients receiving clopidogrel (RR 1.02, 95%CI 0.91 to 1.14; Fig. 2) with low heterogeneity (I2 = 46.7%). Treating study design as a factor in the model yielded no significant interaction with medication (RCTs: RR 0.98, 95%CI 0.87 to 1.11; I2 = 7.8%); observational studies: RR 1.19, 95%CI 0.93 to 1.53; I2 = 65.1%); P = 0.171 for interaction). The null result did not change when studies at serious risk of bias were excluded (RR 1.01, 95%CI 0.91 to 1.14; I2 = 12.2%; n = 10,891) (Figure S2). The GRADE confidence in this estimate is low. No publication bias was found on inspection of funnel plots or on Egger’s regression test.
Major bleeding
Major bleeding was reported in five RCT (n = 5,919) [10, 11, 28, 29, 31] and six observational studies (n = 31,113) [12, 13, 32–35]. Overall, patients receiving low-dose prasugrel exhibited an increased risk of major bleeding relative to patients receiving clopidogrel (RR 1.35, 95%CI 1.10 to 1.67; I2 = 61.4%; n = 37,110; Fig. 3). Separate analysis of observational and RCT studies yielded an increased risk of major bleeding for patients receiving prasugrel in observational studies (RR 1.57, 95%CI 1.23 to 2.01; I2 = 66.3); however, there was no difference in the risk of major bleeding between drug groups in RCTs (RR 0.91, 95%CI 0.61 to 1.36; I2 = 32.1%). To check for possible impact of bias, we repeated the main analysis with studies at serious risk of bias excluded. This analysis replicated the main finding with patients receiving prasugrel exhibiting an increased risk of major bleeding (RR 1.37, 95%CI 1.08 to 1.73; I2 = 64.3%; n = 34,900) (Figure S3). The GRADE confidence in this estimate is low. No publication biases were found either on inspection of funnel plots or on Egger’s regression test.
Minor bleeding
Minor bleeding was reported in five RCTs (n = 5,919) [10, 11, 28, 29, 31] and four observational studies (n = 4,012) [13, 32–34]. Overall, patients receiving low-dose prasugrel exhibited an increased risk of minor bleeding relative to patients receiving clopidogrel (RR 1.47, 95%CI 1.09 to 1.99; I2 53.0%). Separate analysis of observational and RCT studies yielded nonsignificant trends for both types of study design (Figure S4). The GRADE confidence in this estimate is low. No publication bias was found either on inspection of funnel plots or on Egger’s regression test.
Additional outcomes
We found no difference between patients given low-dose prasugrel and standard-dose clopidogrel in the risk of all-cause mortality, cardiovascular death, myocardial infarction, stroke, or stent thrombosis (Figure S5-S9).
Subgroup analyses
Several subgroup analyses were conducted to examine the in different characteristics such as clinical events, aged groups, types of PCI, study country (Table S4). Overall, these subgroups analyses in different characteristics did not change the findings of the main analysis. However, when we separate analyses according to time to measurement outcome, low-dose prasugrel was significant increased risk of major bleeding in short-term (RR 1.97, 95%CI 1.43 to 2.71; I2 = 59.6%) with trended to increase long-term major bleeding (RR 1.14, 95%CI 0.78 to 1.67; I2 = 61.5%) (Figure S10). No publication biases were found either on inspection of funnel plots or on Egger’s regression test in all analyses (Figure S11).
This is the first systematic review and meta-analysis that provides a summary of efficacy and safety of low-dose prasugrel compared to the standard dose of clopidogrel in patients with ACS or patients undergoing PCI. The main findings were as follows: (1) a meta-analysis of RCTs shows that low-dose prasugrel yields similar benefits and risks as the standard dose of clopidogrel as measured by the occurrence of MACE, all-cause mortality, MI, stroke, stent thrombosis, and bleeding; however (2) a meta-analysis of observational studies revealed that the use of low-dose prasugrel was associated with a higher risk of bleeding events compared with clopidogrel, and (3) no higher risk of MACE endpoint events. These findings suggest that low-dose prasugrel offers no benefit relative to clopidogrel in preventing MACE outcomes, but it may be associated with a higher risk of bleeding.
It is well established that DAPT with aspirin and potent P2Y12 inhibitor reduces the risk of major ischemic events significantly in ACS patients undergoing PCI [2, 36, 37]. Previous research has demonstrated that the standard-dose of prasugrel significantly reduces the composite rate of death, MI, and stroke compared with clopidogrel [7, 38]. In addition, the ISAR-REACT 5 trial showed that prasugrel was superior to ticagrelor in reducing the composite end point of death, MI, and stroke at 1 year in patients with ACS who were scheduled to undergo PCI [36]. However, prasugrel leads to an increased likelihood of major bleeding [38–40]. Because of this increased risk of bleeding, the use of standard-dose prasugrel for elderly and low-body weight patients yields no net clinical benefit over standard treatments [1, 2]. In addition, people of Asian heritage exhibit increased rates of bleeding when treated with prasugrel [8, 9]. The higher risk in Asian populations may be due to differences in average body weight, thrombogenicity, or platelet P2Y12 receptor sensitivity [41]. Hence, protocols for the use of prasugrel have adopted lower recommended doses for patients with ACS, particularly in East Asian populations [42]. A large RCT in Japanese patients contributed to these recommendations with the finding that low-dose prasugrel and clopidogrel have similar efficacy and safety profiles [11].
In the present meta-analysis, we examined data collected from patients with ACS or patients undergoing PCI. We found no difference in MACE between patients treated with low-dose prasugrel and patients treated with clopidogrel. This suggests that the advantage in clinical efficacy of prasugrel over clopidogrel is lost when the dose is lowered. However, patients treated with prasugrel still exhibited a higher incidence of major bleeding events. This is generally in agreement with the conclusions drawn in the 12 studies reviewed here. With one exception, all studies used in this analysis reported that prasugrel was associated with either an increase in major bleeding or no effect on major bleeding, and none of the studies used in this analysis reported improvements in the prevalence of MACE outcome events.
As mentioned above, the major motivation to use low-dose as opposed to standard-dose prasugrel has been the need to minimize bleeding risk. This has been a particularly important consideration for Asian patients, who are at higher risk of major bleeding when being treated for ACS or undergoing PCI [43, 44]. Indeed, 10 of the 12 studies we analyzed were done in Japan. Thus, our results suggest that even at a low dose, prasugrel carries an increased risk of bleeding in Japanese patients. It remains for future research to determine whether these effects are similar in other populations.
One confounding issue is that the Japanese protocol for prasugrel use may be different from in other countries. In Japan, most, if not all, P2Y12 inhibitors (low-dose prasugrel or clopidogrel) are administered at the time of diagnosis of ACS before confirmation of the coronary anatomy [13]. This is contrary to ESC guidelines, and likely to be atypical in most locations [3]. One reason for this difference in policy is that some studies in Japan have found advantages for early administration. For example, a recent study in Japan compared prasugrel administration at the time of initial diagnosis to administration after angiography. Patients receiving early administration of prasugrel exhibited reduced rates of in-stent thrombus and plaque protrusion. What’s more, these benefits were not associated with any difference in the rate of bleeding [45]. By contrast, recent studies in other countries have failed to find any benefit for pre-treatment in ACS patients, and have instead found that patients receiving early administration of prasugrel exhibited a substantially higher bleeding risk than patients receiving post-angiography prasugrel [46, 47]. Consideration of the timing of prasugrel administration was beyond the scope of the present meta-analysis. However, the fact that Japanese protocols often indicate early prasugrel administration and the majority of the studies considered here were executed in Japan may be a confounding factor. It may be that more careful consideration of the timing of prasugrel administration could mitigate the bleeding risk reported here.
Another confound in our analysis was that patients in observational studies exhibited an increased risk of bleeding when taking low-dose prasugrel, but those in RCTs did not. It is not clear why this would be, but one issue may be that RCTs may be hesitant to include patients with certain contraindications. How these and other—potentially unmeasured—differences in patient populations may have impacted results is unclear. In general, patients with ACS and undergoing PCI are not very robust, and the observational studies reviewed here may provide more accurate real-world validity than RCTs.
A final confound was that some analyses had high heterogeneity. We noted statistical heterogeneity across the studies in terms of the various study populations, endpoints used and the duration of studies. Some heterogeneity is expected to be reported in meta-analyses of real-world evidence, and complete uniformity can show consistency in bias rather than consistency in real effects. We therefore did a random-effects meta-analysis and several subgroup analyses to explore this heterogeneity, with results in all subgroups that supported the overall conclusions of the study. Relatedly, certain subgroup analyses involved a limited number of patients and should be interpreted with caution.
In general, more studies are necessary to evaluate which risk factors for bleeding are most important for prescribing cardiologists to consider in order to avoid bleeding events. It is possible that prasugrel may be advantageous in patients with cytochrome P450 (CYP) 2C19 poor metabolizer phenotype [48], high platelet reactivity [49], or predisposition towards higher platelet reactivity [50]. A large-scale clinical trial to identify factors which may identify patients who could benefit from the use of prasugrel is warranted.
This is the most comprehensive systematic review to date examining the efficacy and safety of low-dose prasugrel for patients with ACS or undergoing PCI. The use of low-dose prasugrel offers no efficacy improvement relative to clopidogrel, but low-dose prasugrel does carry an increased risk of bleeding complications. These data suggest that prasugrel, even at low doses, should not be used with typical ACS patients or patients undergoing PCI. Many of the studies on which this analysis was based were conducted in Japan with relatively elderly patients. It remains a subject for future research to determine whether prasugrel can be used safely in other populations. Of particular interest would be observational studies replicating the inclusion and exclusion criteria of the Japanese studies reviewed here in other populations.
Acknowledgements
We are grateful to Adam Joseph Osman Dede (Research specialist, School of Pharmaceutical Sciences, University of Phayao) for proofreading and English language review.
Author’s Contributions
Yuttana Wongsalap: design, data collection, analysis, interpretation, writing the manuscript, critical revision, and final approval
Supakorn Ungsriwong: data collection
Wanalee Kumtepm: data collection
Surasak Saokaew: data collection, critical revision, and final approval
Vichai Senthong: critical revision, and final approval
Kirati Kengkla: design, data collection, analysis, interpretation, writing the manuscript, critical revision, and final approval
Funding
The study was partially funded by a grant from University of Phayao under Unit of Excellence on Clinical Outcomes Research and IntegratioN (UNICORN) [grant number: FF64-UoE003].
Data Availability
The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials.
Compliance with Ethical Standards
Ethical approval
This article does not contain any studies with human participants or animals performed by any of the authors.
Conflict of Interest
The authors declare that they have no conflict of interest.