Paracetamol + ibuprofen in children with fever and acute pain

Systematic review and meta-analysis · final figures · v1.0 · 13 Aug 2026

What the data show

0.70
Serious AEs
vs ibuprofen
OR, 95% CI 0.51–0.97 · k=12
0.39
Need for rescue
medication vs paracetamol
OR, 95% CI 0.20–0.75 · k=10
6.0
Odds of being afebrile
at 6 h vs ibuprofen
OR, k=4 — unstable, interval wide

The primary analysis node is dual therapy: paracetamol and ibuprofen given simultaneously or alternating. Regimen is a pre-specified subgroup. Model: random effects (REML) with the Knapp-Hartung adjustment.

PICO

Question 1 (primary) — dual therapy in children

Populationchildren aged 3–18 with fever and/or acute pain. Respiratory infection is the leading clinical context and a subgroup analysis, not an eligibility criterion
Interventionparacetamol + ibuprofen: simultaneously (incl. fixed-dose combination) or alternating
Comparatorparacetamol alone; ibuprofen alone; placebo
Outcomesfever, pain and discomfort, need for rescue medication, tolerability

Question 2 — the fixed-dose combination as a dosage form

Populationany age. Paediatric RCTs of the fixed-dose combination in fever or ARI do not exist: one adenotonsillectomy trial and a phase I study in adolescents are all there is
Interventionfixed-dose combination of ibuprofen + paracetamol in a single dosage form
Comparatormonocomponents, placebo, free combination
Outcomesthe same, plus pharmacokinetics and dose ratio

The answer to the second question is carried over to children as indirect evidence and is labelled as such in every statement. The bridge rests on three supports: the effect of the free combination in children · pharmacokinetic comparability of the fixed and free combination · confirmation of the fixed combination effect in adults.

How the corpus was assembled

RECORDS IDENTIFIED611SCREENED IN AT TITLE/ABSTRACT69INCLUDED IN QUANTITATIVE SYNTHESIS31

Two databases were searched independently: PubMed (474 records) and eLibrary.ru (137). Title/abstract screening was done in duplicate with disagreements resolved; full texts were obtained for 69 records.

How the data were assembled, including the fixed-dose combination

StreamDatabaseFoundSelectedPurpose
APubMed39126main search: dual therapy in children
BPubMed7324fixed-dose combination, all ages
CPubMed53extending the tolerability pool
DeLibrary.ru13710Russian-language layer: Ибуклин data
Tier A — direct bridge (7). Oral FDC with a fever endpoint, in children, or with pharmacokinetic data. The key one is PMID 34304913: the only FDC RCT with temperature as the primary endpoint, full-factorial against each monocomponent and placebo.
Tier B — supportive (6). Oral FDC in adults with acute pain (mostly third-molar extraction) and pooled safety analyses.
Tier C — mentioned only (11). Intravenous forms and chronic or unrelated pain. An intravenous FDC in adult surgery is no bridge to a paediatric dispersible tablet and does not enter the synthesis.

Russian data on Ибуклин. The only source of clinical data on the client's product: PubMed returns zero records for any spelling of the brand. Ten of 137 eLibrary.ru records were selected; four are usable:

  1. Kazan Medical Journal, 2017 — 97 children aged 3–6 with ARI, three arms: paracetamol · ibuprofen · Ибуклин Юниор
    comparative, randomisation not described
    ⚠ republished as eLibrary 28418823 (Praktika Pediatra, 2016) — counted once
    eLibrary 28843727
  2. Doctor.Ru, 2021 — 242 children with fever from acute respiratory infections, Ибуклин Юниор
    prospective single-arm, no comparator
    eLibrary 47302544
  3. Meditsinskiy Sovet, 2022 — 168 children with fever from upper respiratory infections
    observational comparative; no dispersion measures or p-values
    eLibrary 48054910
  4. Farmateka, 2012 — 260 patients: Ибуклин vs Nurofen vs Panadol, three parallel arms; the fever half (n=130) is 87 % respiratory
    randomised, adults aged 18–70
    the eLibrary PDF is corrupt; full text and tables were taken from the journal site
    eLibrary 17804557 · Farmateka 2012;5(238):69–76

These studies do not enter the quantitative synthesis: designs are non-randomised or single-arm, dispersion measures are absent, and some results appear only as figures. They form a separate narrative subsection with a characteristics table and a ROBINS-I risk-of-bias appraisal.

Pooled estimates by comparison

OutcomesEstimate [95% CI]Serious adverse eventsRCTOR 0.92 [0.78, 1.08]Any adverse eventRCTOR 0.92 [0.62, 1.36]Gastrointestinal adverse eventsRCTOR 1.10 [0.85, 1.42]Withdrawal due to adverse eventsRCTOR 1.23 [0.72, 2.08]Need for rescue medicationRCTOR 0.39 [0.20, 0.75]0.30.40.50.60.70.81.01.21.52.02.5← favours interventionfavours comparator →
OutcomesEstimate [95% CI]Serious adverse eventsRCTOR 0.70 [0.51, 0.97]Any adverse eventRCTOR 1.06 [0.93, 1.21]Gastrointestinal adverse eventsRCTOR 0.90 [0.65, 1.24]Withdrawal due to adverse eventsRCTOR 1.52 [1.15, 2.00]Need for rescue medicationRCTOR 0.47 [0.30, 0.75]Afebrile at 6 hoursRCTOR 6.04 [2.06, 17.70]0.30.40.50.60.70.81.01.21.52.02.53.0← favours interventionfavours comparator →
OutcomesEstimate [95% CI]Serious adverse eventsRCTOR 0.32 [0.20, 0.52]Any adverse eventRCTOR 0.67 [0.34, 1.34]Gastrointestinal adverse eventsRCTOR 0.92 [0.20, 4.26]Withdrawal due to adverse eventsRCTOR 0.52 [0.37, 0.73]Need for rescue medicationRCTOR 0.24 [0.13, 0.44]0.30.40.50.60.70.81.01.21.52.02.53.0← favours interventionfavours comparator →

An odds ratio below 1 means the event was less frequent on dual therapy. Marker size is proportional to precision. Statistically significant results are coloured. The number of studies in each pool is given in the row label.

What the pooled estimates are made of

Serious adverse events — dual therapy vs ibuprofenStudyEstimate [95% CI]Erlewyn-Lajeunesse_20067%0.97 [0.02, 50.37]Nabulsi_20067%0.92 [0.02, 47.57]Sarrell_20067%1.00 [0.02, 50.72]Hay_200820%0.32 [0.03, 3.18]Vyas_20147%1.03 [0.02, 53.61]Aitken_201913%0.88 [0.06, 14.22]Daniels_20197%1.01 [0.02, 51.73]Kellstein_20207%0.34 [0.01, 17.23]Motov_20207%1.00 [0.02, 52.04]Smith_20217%1.02 [0.02, 52.25]Su_20217%0.60 [0.01, 30.52]Charde_20257%1.00 [0.02, 51.77]Pooled estimate (k=12)0.70 [0.51, 0.97]0.10.10.20.30.50.71.01.52.03.05.010.020.0← favours dual therapyfavours ibuprofen →I² = 0%
Need for rescue medication — dual therapy vs paracetamolStudyEstimate [95% CI]Viitanen_20036%0.44 [0.18, 1.08]Mehlisch_201028%0.15 [0.10, 0.23]Merry_20107%0.80 [0.33, 1.89]Merry_20131%0.09 [0.01, 0.73]Daniels_201815%0.27 [0.15, 0.49]Daniels_20195%0.21 [0.07, 0.60]Motov_20200%1.00 [0.02, 52.04]Searle_202029%0.32 [0.21, 0.48]Mann_20218%2.17 [1.00, 4.68]Smith_20210%1.01 [0.02, 51.63]Pooled estimate (k=10)0.39 [0.20, 0.75]0.10.10.20.30.50.71.01.52.03.05.010.020.0← favours dual therapyfavours paracetamol →I² = 81%
Need for rescue medication — dual therapy vs ibuprofenStudyEstimate [95% CI]Viitanen_20036%0.34 [0.14, 0.85]Mehlisch_201031%0.74 [0.49, 1.12]Merry_20107%1.29 [0.53, 3.12]Merry_20131%0.09 [0.01, 0.75]Daniels_201816%0.41 [0.23, 0.74]Daniels_20195%0.25 [0.09, 0.68]Motov_20200%1.00 [0.02, 52.04]Searle_202033%0.39 [0.26, 0.59]Smith_20211%0.34 [0.01, 8.41]Pooled estimate (k=9)0.47 [0.30, 0.75]0.10.10.20.30.50.71.01.52.03.05.010.020.0← favours dual therapyfavours ibuprofen →I² = 50%
Serious adverse events — dual therapy vs placeboStudyEstimate [95% CI]Pickering_200237%0.41 [0.07, 2.24]Eccles_20107%0.99 [0.02, 50.39]Daniels_20117%0.33 [0.01, 16.90]Aitken_201922%0.19 [0.02, 1.69]Daniels_20197%0.67 [0.01, 34.26]Kellstein_20207%0.11 [0.00, 5.74]Smith_20217%0.34 [0.01, 17.42]Su_20217%0.22 [0.00, 11.07]Pooled estimate (k=8)0.32 [0.20, 0.52]0.10.10.20.30.50.71.01.52.03.05.010.020.0← favours dual therapyfavours placebo →I² = 0%
Need for rescue medication — dual therapy vs placeboStudyEstimate [95% CI]Pickering_20025%0.28 [0.09, 0.95]Viitanen_200310%0.66 [0.27, 1.62]Mehlisch_201024%0.14 [0.08, 0.25]Atkinson_20157%0.26 [0.09, 0.71]Daniels_201815%0.07 [0.03, 0.15]McHale_20181%0.14 [0.01, 2.94]Daniels_20193%0.12 [0.03, 0.55]Searle_202033%0.77 [0.48, 1.26]Smith_20210%0.34 [0.01, 17.42]You_20261%0.14 [0.01, 1.47]Pooled estimate (k=10)0.24 [0.13, 0.44]0.10.10.20.30.50.71.01.52.03.05.010.020.0← favours dual therapyfavours placebo →I² = 73%
Withdrawal due to adverse events — dual therapy vs placeboStudyEstimate [95% CI]Eccles_20104%0.99 [0.02, 50.39]Daniels_20114%0.33 [0.01, 16.90]Daniels_201811%0.33 [0.03, 3.76]McHale_20184%1.14 [0.02, 59.93]Aitken_201921%0.38 [0.07, 2.08]Daniels_20194%0.67 [0.01, 34.26]Kellstein_20207%0.56 [0.03, 12.02]Smith_202134%0.46 [0.12, 1.77]Su_20217%1.54 [0.08, 29.92]Park_20264%0.98 [0.02, 50.38]Pooled estimate (k=10)0.52 [0.37, 0.73]0.10.10.20.30.50.71.01.52.03.05.010.020.0← favours dual therapyfavours placebo →I² = 0%

Each row is one study; square size is proportional to its weight in the pooled estimate. The diamond is the summary. This shows whether a result rests on several comparable trials or on a single large one. Clicking a row opens the publication.

Where each effect is actually observed

The review answers a paediatric question, but a large part of the evidence comes from adults: fixed-dose combination trials were run mostly in adult post-operative and dental pain. Every significant result is therefore shown with its population split — the pooled estimate alone would overstate what is known about children.

OutcomeAll studiesPopulation split
Serious adverse events
vs ibuprofen
0.70 [0.51; 0.97]
k=12
adults / mixed age: k=5, 0.74 [0.43; 1.28]
children (≤ 18 y): k=7, 0.68 [0.40; 1.16]
Q между подгруппами: p = 0.936
Serious adverse events
vs placebo
0.32 [0.20; 0.52]
k=8
no split — all studies in one population
Afebrile at 6 hours
vs ibuprofen
6.04 [2.06; 17.70]
k=4
no split — all studies in one population
Need for rescue medication
vs ibuprofen
0.47 [0.30; 0.75]
k=9
adults / mixed age: k=6, 0.52 [0.29; 0.92]
children (≤ 18 y): k=3, 0.29 [0.06; 1.49]
Q между подгруппами: p = 0.262
Need for rescue medication
vs paracetamol
0.39 [0.20; 0.75]
k=10
adults / mixed age: k=6, 0.29 [0.15; 0.54]
children (≤ 18 y): k=4, 0.59 [0.07; 5.27]
Q между подгруппами: p = 0.165
Need for rescue medication
vs placebo
0.24 [0.13; 0.44]
k=10
adults / mixed age: k=7, 0.20 [0.09; 0.45]
children (≤ 18 y): k=3, 0.44 [0.10; 2.01]
Q между подгруппами: p = 0.355
Withdrawal due to adverse events
vs ibuprofen
1.52 [1.15; 2.00]
k=11
adults / mixed age: k=9, 1.53 [1.12; 2.10]
children (≤ 18 y): k=2, 1.00 [1.00; 1.00]
Q между подгруппами: p = 0.767
Withdrawal due to adverse events
vs placebo
0.52 [0.37; 0.73]
k=10
no split — all studies in one population

Sensitivity analysis by treatment duration

The product label caps a course at 3–5 days, so a post-hoc sensitivity analysis excluded trials with a documented treatment duration over 7 days. The criterion was applied uniformly to every pool; the only such trial in the entire quantitative synthesis was Doherty 2011 — 13 weeks of dosing for chronic knee pain (osteoarthritis) in patients aged 40–84. Below, every affected pool before and after the restriction. The signal of more frequent withdrawal vs ibuprofen, 68% of whose weight came from that trial, disappears under the short-course restriction; no other conclusion changes.

OutcomeAll studiesShort courses only (≤ 7 days)What changes
Any adverse event
vs paracetamol
0.92 [0.62; 1.36]
k=11, p=0.642
1.01 [0.71; 1.45]
k=10, p=0.947
conclusion unchanged
Any adverse event
vs ibuprofen
1.06 [0.93; 1.21]
k=12, p=0.344
1.08 [0.95; 1.24]
k=11, p=0.215
conclusion unchanged
Withdrawal due to adverse events
vs paracetamol
1.23 [0.72; 2.08]
k=10, p=0.404
0.81 [0.51; 1.28]
k=9, p=0.316
conclusion unchanged
Withdrawal due to adverse events
vs ibuprofen
1.52 [1.15; 2.00]
k=11, p=0.007
1.10 [0.72; 1.67]
k=10, p=0.637
significance disappears
Any adverse event — dual therapy vs paracetamolStudyEstimate [95% CI]Viitanen_20035%1.38 [0.55, 3.46]Erlewyn-Lajeunesse_20060%3.17 [0.12, 80.39]Doherty_20117%0.37 [0.17, 0.81]excludedBaranova_20126%0.91 [0.41, 2.04]Vyas_20141%2.07 [0.35, 12.27]Daniels_201810%0.42 [0.22, 0.78]Aitken_201929%0.72 [0.49, 1.04]Daniels_20199%1.51 [0.77, 2.96]Motov_20200%1.00 [0.02, 52.04]Su_202130%1.30 [0.90, 1.87]Charde_20253%1.59 [0.53, 4.79]All studies (k=11)0.92 [0.62, 1.36]Short courses ≤ 7 d (k=10)1.01 [0.71, 1.45]0.10.10.20.30.50.71.01.52.03.05.010.020.0← favours dual therapyfavours paracetamol →I² = 58%
Any adverse event — dual therapy vs ibuprofenStudyEstimate [95% CI]Viitanen_20034%0.94 [0.39, 2.28]Erlewyn-Lajeunesse_20060%3.00 [0.12, 76.17]Nabulsi_20062%0.73 [0.20, 2.65]Doherty_201110%0.87 [0.49, 1.55]excludedBaranova_20125%1.11 [0.48, 2.57]Vyas_20141%1.43 [0.29, 7.00]Daniels_201811%1.32 [0.77, 2.28]Aitken_201923%1.08 [0.73, 1.59]Daniels_20197%0.70 [0.34, 1.44]Motov_20200%1.00 [0.02, 52.04]Su_202134%1.08 [0.79, 1.48]Charde_20253%1.92 [0.61, 6.01]All studies (k=12)1.06 [0.93, 1.21]Short courses ≤ 7 d (k=11)1.08 [0.95, 1.24]0.10.10.20.30.50.71.01.52.03.05.010.020.0← favours dual therapyfavours ibuprofen →I² = 0%
Withdrawal due to adverse events — dual therapy vs paracetamolStudyEstimate [95% CI]Doherty_201164%2.32 [1.31, 4.12]excludedBaranova_20121%0.80 [0.02, 40.81]Daniels_20184%0.33 [0.03, 3.23]Aitken_20197%0.59 [0.10, 3.54]Daniels_20192%0.33 [0.01, 8.20]Motov_20201%1.00 [0.02, 52.04]Smith_202115%1.01 [0.31, 3.28]Su_20212%3.25 [0.17, 63.04]Charde_20251%1.00 [0.02, 51.77]Park_20261%0.98 [0.02, 50.38]All studies (k=10)1.23 [0.72, 2.08]Short courses ≤ 7 d (k=9)0.81 [0.51, 1.28]0.10.10.20.30.50.71.01.52.03.05.010.020.0← favours dual therapyfavours paracetamol →I² = 19%
Withdrawal due to adverse events — dual therapy vs ibuprofenStudyEstimate [95% CI]Doherty_201168%1.77 [1.05, 2.98]excludedBaranova_20121%0.80 [0.02, 40.81]Daniels_20182%3.08 [0.12, 76.49]Aitken_20195%0.88 [0.12, 6.33]Daniels_20192%0.20 [0.01, 4.18]Kellstein_20202%1.71 [0.08, 35.94]Motov_20201%1.00 [0.02, 52.04]Searle_20202%0.68 [0.04, 10.94]Smith_202112%1.25 [0.37, 4.26]Su_20214%1.82 [0.19, 17.51]Charde_20251%1.00 [0.02, 51.77]All studies (k=11)1.52 [1.15, 2.00]Short courses ≤ 7 d (k=10)1.10 [0.72, 1.67]0.10.10.20.30.50.71.01.52.03.05.010.020.0← favours dual therapyfavours ibuprofen →I² = 0%

In the forests the excluded trial is greyed out and tagged; the two diamonds at the bottom are the all-studies estimate and the short-courses-only estimate. Clicking a row opens the publication.

What to keep in mind

Included studies (31)

  1. Double-blind, placebo-controlled analgesic study of ibuprofen or rofecoxib in combination with paracetamol for tonsillectomy in children. Br J Anaesth. 2002. PMID 11881888 · DOI
  2. Analgesic efficacy of rectal acetaminophen and ibuprofen alone or in combination for paediatric day-case adenoidectomy. Br J Anaesth. 2003. PMID 12925475 · DOI
  3. Alternating ibuprofen and acetaminophen in the treatment of febrile children: a pilot study [ISRCTN30487061]. BMC Med. 2006. PMID 16515705 · DOI
  4. Antipyretic treatment in young children with fever: acetaminophen, ibuprofen, or both alternating in a randomized, double-blind study. Arch Pediatr Adolesc Med. 2006. PMID 16461878 · DOI
  5. Randomised controlled trial of combined paracetamol and ibuprofen for fever. Arch Dis Child. 2006. PMID 16464962 · DOI
  6. Paracetamol plus ibuprofen for the treatment of fever in children (PITCH): randomised controlled trial. BMJ. 2008. PMID 18765450 · DOI
  7. A double-blind, randomised, crossover study of two doses of a single-tablet combination of ibuprofen/paracetamol and placebo for primary dysmenorrhoea. Curr Med Res Opin. 2010. PMID 20942614 · DOI
  8. Efficacy of standard doses of Ibuprofen alone, alternating, and combined with acetaminophen for the treatment of febrile children. Clin Ther. 2010. PMID 21353111 · DOI
  9. A single-tablet fixed-dose combination of racemic ibuprofen/paracetamol in the management of moderate to severe postoperative dental pain in adult and adolescent patients: a multicenter, two-stage, randomized, double-blind, parallel-group, placebo-controlled, factorial study. Clin Ther. 2010. PMID 20637958 · DOI
  10. Combined acetaminophen and ibuprofen for pain relief after oral surgery in adults: a randomized controlled trial. Br J Anaesth. 2010. PMID 20007794 · DOI
  11. A randomised controlled trial of ibuprofen, paracetamol or a combination tablet of ibuprofen/paracetamol in community-derived people with knee pain. Ann Rheum Dis. 2011. PMID 21804100 · DOI
  12. A randomised, five-parallel-group, placebo-controlled trial comparing the efficacy and tolerability of analgesic combinations including a novel single-tablet combination of ibuprofen/paracetamol for postoperative dental pain. Pain. 2011. PMID 21257263 · DOI
  13. Эффективность и безопасность фиксированной комбинации ибупрофен/парацетамол при лихорадочном и болевом синдромах в амбулаторной практике. Фарматека. 2012. eLibrary 17804557
  14. Ibuprofen, paracetamol, and steam for patients with respiratory tract infections in primary care: pragmatic randomised factorial trial. BMJ. 2013. PMID 24162940 · DOI
  15. Randomized comparison between the combination of acetaminophen and ibuprofen and each constituent alone for analgesia following tonsillectomy in children. Can J Anaesth. 2013. PMID 24150660 · DOI
  16. Randomized comparative trial of efficacy of paracetamol, ibuprofen and paracetamol-ibuprofen combination for treatment of febrile children. Perspect Clin Res. 2014. PMID 24551584 · DOI
  17. Combination paracetamol and ibuprofen for pain relief after oral surgery: a dose ranging study. Eur J Clin Pharmacol. 2015. PMID 25778933 · DOI
  18. Alternating Acetaminophen and Ibuprofen versus Monotherapies in Improvements of Distress and Reducing Refractory Fever in Febrile Children: A Randomized Controlled Trial. Paediatr Drugs. 2017. PMID 28523589 · DOI
  19. Analgesic Efficacy of an Acetaminophen/Ibuprofen Fixed-dose Combination in Moderate to Severe Postoperative Dental Pain: A Randomized, Double-blind, Parallel-group, Placebo-controlled Trial. Clin Ther. 2018. PMID 30245281 · DOI
  20. Do children undergoing bilateral myringotomy with placement of ventilating tubes benefit from pre-operative analgesia? A double-blinded, randomised, placebo-controlled trial. J Laryngol Otol. 2018. PMID 29998815 · DOI
  21. An integrated safety analysis of combined acetaminophen and ibuprofen (Maxigesic ® /Combogesic®) in adults. J Pain Res. 2019. PMID 30804681 · DOI
  22. Efficacy and Safety of an Intravenous Acetaminophen/Ibuprofen Fixed-dose Combination After Bunionectomy: a Randomized, Double-blind, Factorial, Placebo-controlled Trial. Clin Ther. 2019. PMID 31447129 · DOI
  23. Comparison of Oral Ibuprofen and Acetaminophen with Either Analgesic Alone for Pediatric Emergency Department Patients with Acute Pain. J Emerg Med. 2020. PMID 32247660 · DOI
  24. Evaluation of Fixed-Dose Combinations of Ibuprofen and Acetaminophen in the Treatment of Postsurgical Dental Pain: A Pilot, Dose-Ranging, Randomized Study. Drugs R D. 2020. PMID 32506309 · DOI
  25. Efficacy and Safety of Single and Multiple Doses of a Fixed-dose Combination of Ibuprofen and Acetaminophen in the Treatment of Postsurgical Dental Pain: Results From 2 Phase 3, Randomized, Parallel-group, Double-blind, Placebo-controlled Studies. Clin J Pain. 2020. PMID 32271183 · DOI
  26. Opioid-free anesthesia for adenotonsillectomy in children. Int J Pediatr Otorhinolaryngol. 2021. PMID 33290925 · DOI
  27. Safety and tolerability of fixed-dose combinations of ibuprofen and acetaminophen: pooled analysis of phase 1-3 clinical trials. Postgrad Med. 2021. PMID 33821768 · DOI
  28. Efficacy of a Fixed-Dose Combination of Ibuprofen and Acetaminophen Compared With Individual Monocomponents in Adult Male Subjects With Endotoxin-Induced Fever: A Randomized Controlled Trial. Clin Ther. 2021. PMID 34304913 · DOI
  29. Assessing the safety and adverse effects of paracetamol, ibuprofen, and their combination in paediatric pain and fever management: A prospective observational study. J Family Med Prim Care. 2025. PMID 40115566 · DOI
  30. Effect of an Intravenous Acetaminophen/Ibuprofen Fixed-Dose Combination on Catheter-Related Bladder Discomfort: A Prospective, Randomized, Placebo-Controlled, Double-Blind Pilot Study. Medicina (Kaunas). 2026. PMID 42356051 · DOI
  31. Efficacy of an intravenous acetaminophen/ibuprofen fixed-dose combination for postoperative catheter-related bladder discomfort in patients undergoing robot-assisted laparoscopic prostatectomy: a double-blinded, randomized, placebo-controlled study. Korean J Anesthesiol. 2026. PMID 41918243 · DOI