Outcomes for patients with COVID-19 admitted to Australian intensive care units during the first four months of the pandemic.

OBJECTIVES: To describe the characteristics and outcomes of patients with COVID-19 admitted to intensive care units (ICUs) during the initial months of the pandemic in Australia. DESIGN, SETTING: Prospective, observational cohort study in 77 ICUs across Australia. PARTICIPANTS: Patients admitted to participating ICUs with laboratory-confirmed COVID-19 during 27 February - 30 June 2020. MAIN OUTCOME MEASURES: ICU mortality and resource use (ICU length of stay, peak bed occupancy). RESULTS: The median age of the 204 patients with COVID-19 admitted to intensive care was 63.5 years (IQR, 53-72 years); 140 were men (69%). The most frequent comorbid conditions were obesity (40% of patients), diabetes (28%), hypertension treated with angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers (24%), and chronic cardiac disease (20%); 73 patients (36%) reported no comorbidity. The most frequent source of infection was overseas travel (114 patients, 56%). Median peak ICU bed occupancy was 14% (IQR, 9-16%). Invasive ventilation was provided for 119 patients (58%). Median length of ICU stay was greater for invasively ventilated patients than for non-ventilated patients (16 days; IQR, 9-28 days v 3 days; IQR, 2-5 days), as was ICU mortality (26 deaths, 22%; 95% CI, 15-31% v four deaths, 5%; 95% CI, 1-12%). Higher Acute Physiology and Chronic Health Evaluation II (APACHE-II) scores on ICU day 1 (adjusted hazard ratio [aHR], 1.15; 95% CI, 1.09-1.21) and chronic cardiac disease (aHR, 3.38; 95% CI, 1.46-7.83) were each associated with higher ICU mortality. CONCLUSION: Until the end of June 2020, mortality among patients with COVID-19 who required invasive ventilation in Australian ICUs was lower and their ICU stay longer than reported overseas. Our findings highlight the importance of ensuring adequate local ICU capacity, particularly as the pandemic has not yet ended.

Calcium desensitisation in late polymicrobial sepsis is associated with loss of vasopressor sensitivity in a murine model.

BACKGROUND: Sepsis is characterised by diminished vasopressor responsiveness. Vasoconstriction depends upon a balance: Ca(2+)-dependent myosin light-chain kinase promotes and Ca(2+)-independent myosin light-chain phosphatase (MLCP) opposes vascular smooth muscle contraction. The enzyme Rho kinase (ROK) inhibits MLCP, favouring vasoconstriction. We tested the hypothesis that ROK-dependent MLCP inhibition was attenuated in late sepsis and associated with reduced contractile responses to certain vasopressor agents. METHODS: This is a prospective, controlled animal study. Sixteen-week-old C57/BL6 mice received laparotomy or laparotomy with caecal ligation and puncture (CLP). Antibiotics, fluids and analgesia were provided before sacrifice on day 5. Vasoconstriction of the femoral arteries to a range of stimuli was assessed using myography: (i) depolarisation with 87 mM K(+) assessed voltage-gated Ca(2+) channels (L-type, Cav1.2 Ca(2+) channels (LTCC)), (ii) thromboxane A2 receptor activation assessed the activation state of the LTCC and ROK/MLCP axis, (iii) direct PKC activation (phorbol-dibutyrate (PDBu), 5 µM) assessed the PKC/CPI-17 axis independent of Ca(2+) entry and (iv) α1-adrenoceptor stimulation with phenylephrine (10(-8) to 10(-4) M) and noradrenaline (10(-8) to 10(-4) M) assessed the sum of these pathways plus the role of the sarcoplasmic reticulum (SR). ROK-dependent MLCP activity was indexed by Western blot analysis of P[Thr855]MYPT. Parametric and non-parametric data were analysed using unpaired Student's t-tests and Mann-Whitney tests, respectively. RESULTS: ROK-dependent inhibition of MLCP activity was attenuated in both unstimulated (n = 6 to 7) and stimulated (n = 8 to 12) vessels from mice that had undergone CLP (p < 0.05). Vessels from CLP mice demonstrated reduced vasoconstriction to K(+), thromboxane A2 receptor activation and PKC activation (n = 8 to 13; p < 0.05). α1-adrenergic responses were unchanged (n = 7 to 12). CONCLUSIONS: In a murine model of sepsis, ROK-dependent inhibition of MLCP activity in vessels from septic mice was reduced. Responses to K(+) depolarisation, thromboxane A2 receptor activation and PKC activation were diminished in vitro whilst α1-adrenergic responses remained intact. Inhibiting MLCP may present a novel therapeutic target to manage sepsis-induced vascular dysfunction.

Attenuated platelet aggregation in patients with septic shock is independent from the activity state of myosin light chain phosphorylation or a reduction in Rho kinase-dependent inhibition of myosin light chain phosphatase.

BACKGROUND: Impaired coagulation contributes to the morbidity and mortality associated with septic shock. Whether abnormal platelet contraction adds to the bleeding tendency is unknown. Platelets contract when Ca(2+)-dependent myosin light chain kinase (MLCK) phosphorylates Ser19 of myosin light chain (MLC20), promoting actin-myosin cross-bridge cycling. Contraction is opposed when myosin light chain phosphatase (MLCP) dephosphorylates MLC20. It is thought that Rho kinase (ROK) inhibits MLCP by phosphorylating Thr855 of the regulatory subunit MYPT, favouring platelet contraction. This study tested the hypotheses that in septic shock, (i) platelet function is inversely correlated with illness severity and (ii) ROK-dependent MLCP inhibition and myosin light chain phosphorylation are reduced. METHODS: Blood was sampled from non-septic shock patients and patients in the first 24 h of septic shock. Platelet function was assessed using whole blood impedance aggregation induced by 1) ADP (1.6 and 6.5 µM), 2) thrombin receptor-activating protein (TRAP; 32 µM), 3) arachidonic acid (500 µM) and 4) collagen (3.2 µg/ml). Arachidonic acid-induced aggregation was measured in the presence of the ROK inhibitor Y27632. Illness severity was evaluated using sequential organ failure assessment (SOFA) and acute physiology and chronic health evaluation (APACHE) II scores. Western blot analysis of [Ser19]MLC20 and [Thr855]MYPT phosphorylation quantified activation and inhibition of platelet MLC20 and MLCP, respectively. Data were analysed using Spearman's rank correlation coefficient, Student's t-test and Mann-Whitney test; p < 0.05 was considered significant. RESULTS: Agonist-induced aggregation was attenuated in septic shock patients (n = 22 to 34; p < 0.05). Aggregation correlated inversely with SOFA and APACHE II scores (n = 34; p < 0.05). Thr855 phosphorylation of MYPT from unstimulated platelets was not decreased in patients with septic shock (n = 22 to 24). Both septic shock and ROK inhibition attenuated arachidonic acid-induced platelet aggregation independent of changes in [Ser19]MLC20 and [Thr855]MYPT phosphorylation (n = 14). CONCLUSIONS: Impairment of whole blood aggregation in patients within the first 24 h of septic shock was correlated with SOFA and APACHE II scores. Attenuated aggregation was independent of molecular evidence of diminished platelet contraction or reduced ROK inhibition of MLCP. Efforts to restore platelet function in septic shock should therefore focus on platelet adhesion and degranulation.