Machine learning approaches for practical predicting outpatient near-future AECOPD based on nationwide electronic medical records.

In this research, we aimed to harness machine learning to predict the imminent risk of acute exacerbation in chronic obstructive pulmonary disease (AECOPD) patients. Utilizing retrospective data from electronic medical records of two Taiwanese hospitals, we identified 26 critical features. To predict 3- and 6-month AECOPD occurrences, we deployed five distinct machine learning algorithms alongside ensemble learning. The 3-month risk prediction was best realized by the XGBoost model, achieving an AUC of 0.795, whereas the XGBoost was superior for the 6-month prediction with an AUC of 0.813. We conducted an explainability analysis and found that the episode of AECOPD, mMRC score, CAT score, respiratory rate, and the use of inhaled corticosteroids were the most impactful features. Notably, our approach surpassed predictions that relied solely on CAT or mMRC scores. Accordingly, we designed an interactive prediction system that provides physicians with a practical tool to predict near-term AECOPD risk in outpatients.

Synergistic interaction of guanfacine or dexmedetomidine coadministered with lidocaine for cutaneous analgesia in rats.

BACKGROUND: This study examined the cutaneous analgesic effects of lidocaine co-injected with guanfacine and its comparison with dexmedetomidine. METHODS: Cutaneous analgesic effects are quantified through the blocking effects of the cutaneous trunci muscle reflex against skin pinpricks in rats. The dose-response curves of guanfacine, dexmedetomidine, and lidocaine were constructed and drug-drug interactions were analyzed by the ED50 isobologram. RESULTS: Subcutaneous injections of guanfacine, dexmedetomidine, and lidocaine produced dose-dependently nociceptive/sensory blockade. On the ED50 (50% effective dose) basis, the potency rankings of the drug are dexmedetomidine (0.09 [0.08-0.11] µmol/kg) > guanfacine (3.98 [2.96-5.34] µmol/kg) > lidocaine (25.40 [23.51-27.44] µmol/kg) (p < 0.01). On their equipotent doses (ED25, ED50, and ED75), the duration of sensory blockade induced by guanfacine or dexmedetomidine was longer than lidocaine's (p < 0.01). Both guanfacine and dexmedetomidine showed synergistic effects with lidocaine. CONCLUSIONS: We showed that guanfacine elicits dose-dependent cutaneous analgesia when administered subcutaneously. Lidocaine is less potent than guanfacine or dexmedetomidine. Both guanfacine and dexmedetomidine enhance the potency and duration of lidocaine. Better synergistic responses we are getting with guanfacine plus lidocaine.

Intrathecal dopamine and serotonin enhance motor and nociceptive blockades of lidocaine in rats.

The study examined the effect of intrathecal injection of dopamine (serotonin) and/or lidocaine. Intrathecal injections of dopamine (serotonin or epinephrine), lidocaine, or their combination were carried out in male Sprague Dawley rats. Neurobehavioral examinations (motor and nociceptive reactions) were performed before and after spinal injection. Intrathecal serotonin (1.5 µmol), dopamine (2.5 µmol), epinephrine (1:40000), and lidocaine (0.75 µmol) produced 29%, 33%, 29%, and 54% nociceptive blockade, whereas serotonin (1.5 µmol), dopamine (2.5 µmol), or epinephrine (1:40000) produced a longer duration of nociceptive blockade than lidocaine (0.75 µmol) (P < 0.05). Serotonin (1.5 µmol), dopamine (1.25 and 2.5 µmol), or epinephrine (1:40000 and 1:80000) prolonged the duration and increased the potency of spinal motor and nociceptive blockades of lidocaine (50% effective dose, ED50) (P < 0.05). The motor and nociceptive blockades caused by lidocaine (ED50) plus dopamine (2.5 µmol) or lidocaine (ED50) plus epinephrine (1:40000) were more outstanding than lidocaine (ED50) plus serotonin (0.75 µmol) (P < 0.05). Our study provides evidence that intrathecal dopamine or serotonin produces spinal nociceptive blockade dose-dependently. Dopamine and serotonin are less potent than lidocaine in inducing spinal nociceptive blockade. When mixed with lidocaine solution, dopamine or serotonin improves spinal motor and nociceptive blockades. The motor and nociceptive blockade caused by lidocaine (ED50) plus dopamine (2.5 µmol) is similar to that caused by lidocaine (ED50) plus epinephrine (1:40000).

Quality of life and cost-effectiveness of different breast cancer surgery procedures: a Markov decision tree-based approach in the framework of Predictive, Preventive, and Personalized Medicine.

Purpose: Breast cancer is a complex disease with heterogeneous outcomes that may benefit from the implementation of Predictive, Preventive, and Personalized Medicine (PPPM/3PM) strategies. In this study, we aimed to explore the potential of PPPM approaches by investigating the 10-year trends in quality of life (QOL) and the cost-effectiveness of different types of surgeries for patients with breast cancer. Methods: This prospective cohort study recruited 144 patients undergoing breast conserving surgery (BCS), 199 undergoing modified radical mastectomy (MRM), and 44 undergoing total mastectomy with transverse rectus abdominis myocutaneous flap (TRAMF) from three medical centers in Taiwan between June 2007 and June 2010. Results: All patients exhibited a significant decrease in most QOL dimension scores from before surgery to 6 months postoperatively (p < 0.05); however, from postoperative year 1 to 2, improvement in most QOL dimension scores was significantly better in the TRAMF group than in the BCS and MRM groups (p < 0.05). At 2, 5, and 10 years after surgery, the patients' QOL remained stable. In the Markov decision tree model, the TRAMF group had higher total direct medical costs than the MRM and BCS groups (US$ 32,426, US$ 29,487, and US$ 28,561, respectively) and higher average QALYs gained (7.771, 6.773, and 7.385, respectively), with an incremental cost-utility ratio (ICUR) of US$ 2,944.39 and US$ 10,013.86 per QALY gained. Conclusions: TRAMF appeared cost effective compared with BCS and MRM, and it has been proved with considerable QOL improvements in the framework of PPPM. Future studies should continue to explore the potential of PPPM approaches in breast cancer care. By incorporating predictive models, personalized treatment plans, and preventive strategies into routine clinical practice, we can further optimize patient outcomes and reduce healthcare costs associated with breast cancer treatment. Supplementary Information: The online version contains supplementary material available at 10.1007/s13167-023-00326-4.

Medium-term surgical outcomes and health-related quality of life after laparoscopic vs open colorectal cancer resection: SF-36 health survey questionnaire.

BACKGROUND: Previous studies that compared the postoperative health-related quality of life (HRQoL) outcomes after receiving laparoscopic resection (LR) or open resection (OR) in patients with colorectal cancer (CRC) have different conclusions. AIM: To explore the medium-term effect of postoperative HRQoL in such patients. METHODS: This study randomized 567 patients undergoing non-metastatic CRC surgery managed by one surgeon to the LR or OR groups. HRQoL was assessed during the preoperative period and 3, 6, and 12 mo postoperative using a modified version of the 36-Item Short Form (SF-36) Health Survey questionnaire, emphasizing eight specific items. RESULTS: This cohort randomly assigned 541 patients to receive LR (n = 296) or OR (n = 245) surgical procedures. More episodes of postoperative urinary tract infection (P < 0.001), wound infection (P < 0.001), and pneumonia (P = 0.048) were encountered in the OR group. The results demonstrated that the LR group subjectively gained mildly better general health (P = 0.045), moderately better physical activity (P = 0.006), and significantly better social function recovery (P = 0.0001) 3 mo postoperatively. Only the aspect of social function recovery was claimed at 6 mo, with a significant advantage in the LR group (P = 0.001). No clinical difference was found in HRQoL during the 12 mo. CONCLUSION: Our results demonstrated that LR resulted in better outcomes, including intra-operative blood loss, surgery-related complications, course of recovery, and especially some health domains of HRQoL at least within 6 mo postoperatively. Patients should undergo LR if there is no contraindication.

Impact of intravenous vitamin C as a monotherapy on mortality risk in critically ill patients: A meta-analysis of randomized controlled trials with trial sequential analysis.

Background: This meta-analysis aimed at investigating the pooled evidence regarding the effects of intravenous vitamin C (IVVC) on mortality rate in critically ill patients. Methods: Databases including Medline, Embase, and Cochrane Library were searched from inception to October, 2022 to identify RCTs. The primary outcome was the risk of overall mortality. Subgroup analyses were performed based on IVVC dosage (i.e., cut-off value: 100 mg/kg/day or 10000 mg/day). Trial sequential analysis (TSA) was used to examine the robustness of evidence. Results: A total of 12 trials including 1,712 patients were analyzed. Although meta-analysis demonstrated a lower risk of mortality in patients with IVVC treatment compared to those without [risk ratio (RR): 0.76, 95% CI: 0.6 to 0.97, p = 0.02, I 2 = 36%, 1,711 patients), TSA suggested the need for more studies for verification. Moreover, subgroup analyses revealed a reduced mortality risk associated with a low IVVC dosage (RR = 0.72, p = 0.03, 546 patients), while no beneficial effect was noted with high IVVC dosage (RR = 0.74, p = 0.13, I 2 = 60%, 1,165 patients). The durations of vasopressor [mean difference (MD): -37.75 h, 404 patients) and mechanical ventilation (MD: -47.29 h, 388 patients) use were shorter in the IVVC group than those in the controls, while there was no significant difference in other prognostic outcomes (e.g., length of stay in intensive care unit/hospital) between the two groups. Conclusion: Although intravenous vitamin C as a monotherapy reduced pooled mortality, durations of vasopressor use and mechanical ventilation, further research is required to support our findings and to identify the optimal dosage of vitamin C in the critical care setting. Systematic review registration: https://www.crd.york.ac.uk/prospero/, identifier CRD42022371090.

Association of a low vitamin D status with risk of post-stroke depression: A meta-analysis and systematic review.

Background: Although post-stroke depression (PSD) affects one-third of patients following an acute stroke, pooled evidence addressing the correlation between a low vitamin D status and the risk of PSD remains inconclusive. Methods: Comprehensive database search of Medline, EMBASE, Cochrane library, and Google Scholar was performed from inception to December 2022. The primary outcome was the association of PSD risk with a low vitamin D status, while the secondary outcomes included the relationship between PSD and other risk factors. Results: Analysis of seven observational studies published between 2014 and 2022 with 1,580 patients showed pooled incidences of vitamin D deficiency (defined as 25[OH] D levels < 50 nmol/L) and PSD of 60.1 and 26.1%, respectively. Patients with PSD had a lower circulating vitamin D concentration compared to those without [mean difference (MD) =-13.94 nmol/L, 95% CI: -21.83 to -6.05, p = 0.0005, I 2 = 91%, six studies, 1,414 patients]. Meta-analysis also demonstrated a correlation between a low vitamin D level and an increased PSD risk [odd ratio (OR) = 3.25, 95% CI: 1.57-6.69, p = 0.001, I 2 = 78.7%, 1,108 patients], the heterogeneity of which was found to be associated with the incidence of vitamin D deficiency but not female proportion on meta-regression. Besides, female gender (OR = 1.78, 95% CI: 1.3-2.44, p = 0.003, I 2 = 31%, five studies, 1,220 patients), hyperlipidemia (OR = 1.55, 95% CI: 1.01-2.36, p = 0.04, I 2 = 0%, four studies, 976 patients), and high National Institutes of Health Stroke Scale (NIHSS) scores (MD = 1.45, 95% CI: 0.58-2.32, p = 0.001, I 2 = 82%, five studies, 1,220 patients) were potential risk factors for PSD. For the primary outcome, the certainty of evidence was very low. Regarding secondary outcomes, the certainty of evidence was low for BMI, female gender, hypertension, diabetes, and stroke history, and very low for age, level of education, hyperlipidemia, cardiovascular disease, and NIHSS scores. Conclusion: The results suggested an association of a low circulating vitamin D level with an increased risk of PSD. Besides, female gender, hyperlipidemia, high NIHSS score were related to an increased risk or occurrence of PSD. The current study may imply the necessity of routine circulating vitamin D screening in this population. Systematic review registration: https://www.crd.york.ac.uk/prospero/, identifier CRD42022381580.