Daily Respiratory Research Analysis
Analyzed 199 papers and selected 3 impactful papers.
Summary
A randomized crossover trial shows that chronic SSRI therapy (paroxetine or escitalopram) blunts the hypercapnic ventilatory response and further depresses ventilation when co-administered with oxycodone, highlighting a clinically relevant safety signal. A Nature Biomedical Engineering study introduces a “tripod-like” lung-targeting lipid nanoparticle platform that achieves >90% lung selectivity and markedly boosts mRNA and CRISPR-Cas9 delivery. A Science Advances analysis using U.S. death certificates estimates ~155,000 unrecognized COVID-19 deaths, revealing inequitable undercounting across disadvantaged populations.
Research Themes
- Respiratory pharmacovigilance and opioid safety
- Lung-targeted gene delivery and genome editing
- Pandemic mortality surveillance and health equity
Selected Articles
1. Effect of Paroxetine or Escitalopram Co-administered with Oxycodone vs Oxycodone Alone on Ventilation During Hypercapnia: A Randomized Clinical Trial.
In a double-blind, randomized 3-period crossover trial in healthy adults, 21 days of paroxetine or escitalopram significantly reduced hypercapnic ventilatory response versus placebo and further depressed ventilation when co-administered with oxycodone. The consistent effect across two SSRIs supports a class effect on ventilatory control with chronic use.
Impact: This trial identifies a reproducible, clinically relevant respiratory safety signal for commonly used SSRIs, especially pertinent amid widespread opioid prescribing. It has immediate implications for co-prescribing and monitoring in patients at risk of respiratory depression.
Clinical Implications: Avoid or carefully monitor SSRI–opioid co-prescribing in patients with respiratory risk (e.g., sleep-disordered breathing, COPD, elderly). Consider dose adjustments, alternative analgesics, and counseling about sedation/respiratory depression; incorporate ventilatory risk into perioperative and pain management plans.
Key Findings
- After 21 days, paroxetine + oxycodone reduced hyperoxic–hypercapnic ventilation by −6.5 L/min vs oxycodone alone (P<0.001).
- Escitalopram + oxycodone reduced ventilation by −5.5 L/min vs oxycodone alone on day 21 (P=0.001).
- SSRIs alone decreased ventilation by ~−6.5 to −6.9 L/min vs placebo on day 20, indicating a class effect with chronic use.
Methodological Strengths
- Randomized, double-blind, 3-period crossover design with within-subject controls.
- Standardized measurement of hypercapnic ventilatory response using Duffin’s rebreathing method.
Limitations
- Small sample of healthy volunteers; limited generalizability to patients with comorbidities.
- Hyperoxic testing conditions and surrogate physiological endpoint; no clinical outcome data.
Future Directions: Replicate in patient populations (chronic pain, depression, OSA/COPD), evaluate other SSRIs/SNRIs and dose–response, and integrate pharmacovigilance/EMR studies to quantify clinical respiratory events.
BACKGROUND: Opioid-induced respiratory depression remains a critical public safety concern. Prior clinical findings demonstrated decreased hypercapnic ventilation after 5 days when paroxetine, a Selective Serotonin Reuptake Inhibitor (SSRI), was administered alone or with oxycodone. However, uncertainty remained whether chronic use of SSRIs could cause similar respiratory effects. This study investigated whether chronic use of paroxetine and another SSRI, escitalopram, lead to a similar decrease in ventilatory response to hypercapnia. METHOD: In this randomized, double-blind, 3-period crossover trial, healthy participants were administered one of the following: paroxetine 40 mg from days 1 to 6 and 60 mg from days 7 to 21, escitalopram 20 mg from days 1 to 6 and 30 mg from days 7 to 21, and placebo from days 1 to 21. Oxycodone 10 mg was co-administered on days 6, 12, and 21. Hyperoxic-hypercapnic ventilation was measured using Duffin's rebreathing. RESULTS: Of the 27 participants, 22 (81%) completed the trial. Paroxetine and escitalopram both significantly decreased hyperoxic-hypercapnic ventilation when co-administered with oxycodone compared to oxycodone alone on day 21 (paroxetine mean difference, -6.5 L/min [1-sided 97.5% CI, -∞ to -3.1] P<0.001, escitalopram mean difference, -5.5 L/min [1-sided 97.5% CI, -∞ to -2.1] P=0.001) and when administered alone compared to placebo on day 20 (paroxetine mean difference, -6.5 L/min [1-sided 97.5% CI, -∞ to -2.1] P=0.003, escitalopram mean difference, -6.9 L/min [1-sided 97.5% CI, -∞ to -2.5] P=0.002). CONCLUSIONS: Both paroxetine and escitalopram, alone and co-administered with oxycodone, decrease hypercapnic ventilation after 21 days suggesting that selective serotonin reuptake inhibitors may have a class effect on hypercapnic ventilation that persists with chronic use.
2. Applying machine learning to identify unrecognized COVID-19 deaths recorded as other causes of death in the United States.
Using machine learning on U.S. death certificates (Mar 2020–Dec 2021), the authors estimate ~155,536 unrecognized COVID-19 deaths, implying that official counts underreported mortality by ~19%. Undercounting was concentrated among disadvantaged groups and regions, revealing hidden inequities in the death investigation system.
Impact: Quantifying hidden pandemic mortality reshapes respiratory pandemic situational awareness and informs policy on surveillance, certification, and resource allocation, especially for health equity.
Clinical Implications: Enhance death certification training, integrate WBE and model-based surveillance, and prioritize targeted outreach in undercounted communities; adjust risk communication and healthcare capacity planning to reflect true respiratory mortality burden.
Key Findings
- Machine learning predicted 155,536 (95% UI: 150,062–161,112) unrecognized COVID-19 deaths in Mar 2020–Dec 2021.
- Estimated U.S. COVID-19 mortality was 19% higher than official counts.
- Undercounting was concentrated among less-educated decedents, racial/ethnic minorities, low-income/worse-health counties, and the South.
Methodological Strengths
- National-scale analysis using individual-level death certificates with machine learning and uncertainty quantification.
- Equity-focused stratification across education, race/ethnicity, socioeconomic and regional factors.
Limitations
- Relies on the accuracy and completeness of recorded variables; lacks gold-standard cause-of-death adjudication.
- Generalizability to later pandemic phases (e.g., post-vaccination) may differ.
Future Directions: Extend models to later waves, integrate healthcare utilization and WBE data, and evaluate structural reforms that reduce certification bias and inequities.
The actual number of US deaths caused by severe acute respiratory syndrome coronavirus 2 infection has been investigated and debated since the start of the COVID-19 pandemic. Here, we use machine learning trained on US death certificates from March 2020 to December 2021 to predict 155,536 (95% uncertainty interval: 150,062 to 161,112) unrecognized COVID-19 deaths. This indicates that 19% more COVID-19 deaths occurred in the US than officially reported. Predicted unrecognized COVID-19 deaths occurred disproportionately among decedents with less than a high school education; decedents identified as Hispanic, American Indian, Alaska Native, Asian, and/or Black; counties with lower household incomes and worse preexisting health; and counties in the South. These findings suggest that the US death investigation system undercounted COVID-19 deaths unevenly, hiding the true extent of inequities.
3. 'Tripod-like' lung-targeting (LuT) lipids for highly efficient and selective LNPs for gene delivery and editing.
A 444-lipid screen identified a ‘tripod-like’ LuT motif that enables lung-selective LNPs (>90%) with markedly enhanced mRNA delivery (25.5×) and CRISPR editing (9.2×) versus DOTAP SORT benchmarks. The lead 1A7B13 LNPs improved endosomal escape and demonstrated therapeutic efficacy delivering IL-10 mRNA in acute lung injury models.
Impact: This platform addresses a long-standing delivery barrier by achieving efficient, lung-selective nucleic acid delivery/editing, opening translational avenues for cystic fibrosis, surfactant disorders, pulmonary fibrosis, and lung cancer.
Clinical Implications: While preclinical, lung-selective LNPs could enable lower doses, reduced off-target toxicity, and inhaled or systemic delivery of mRNA/CRISPR therapeutics for monogenic and inflammatory lung diseases.
Key Findings
- A ‘tripod-like’ quaternary amine head with three long alkyl chains was the optimal motif among 444 LuT lipids.
- Lead 1A7B13 LNPs achieved 25.5-fold higher mRNA delivery and 9.2-fold higher CRISPR editing versus DOTAP SORT, with >90% lung selectivity.
- Improved endosomal escape and favorable plasma protein adsorption underpinned enhanced delivery; IL-10 mRNA conferred benefit in acute lung injury.
Methodological Strengths
- Large structure–activity screen with in vivo validation across mRNA and CRISPR cargos.
- Head-to-head benchmarking versus a recognized lung-targeting LNP (DOTAP SORT) and mechanistic analyses (endosomal escape, protein corona).
Limitations
- Preclinical animal studies; long-term safety, immunogenicity, and manufacturability remain to be established.
- Potential species-specific biodistribution may affect human translation.
Future Directions: Toxicology and biodistribution in large animals, inhalation formulations, GMP manufacturing, and efficacy in monogenic lung disease models (e.g., CFTR, SFTPC).
Developing lung-targeting delivery systems is essential for treating pulmonary conditions such as genetic respiratory diseases, infections, fibrosis and cancer. We synthesized and evaluated 444 lung-targeting lipids (LuT lipids) that form lipid nanoparticles (LNPs) to efficiently deliver messenger RNA and CRISPR-Cas9 genome editors to lungs with minimal side effects. Empirical analyses revealed structure-activity relationships, with top-performing LuT lipids possessing a unique 'tripod-like' structure consisting of a quaternary amine head, three long alkyl chains as legs and a short chain as a handle. LuT lipids improved endosomal escape, cargo release and endogenous targeting via adsorption of plasma proteins. Lead 1A7B13 LNPs showed a 25.5-fold improvement in mRNA delivery and a 9.2-fold increase in CRISPR-Cas9 gene-editing efficiency compared to benchmark DOTAP SORT LNPs, achieving over 90% selectivity to the lungs. 1A7B13 LNPs effectively delivered IL-10 mRNA in a therapeutic model of acute lung injury. This study reveals the relationship between lipid structure and lung-targeting activity, enriching the toolkit for lung-specific carriers.