Characterisation of cardiovascular function in intensive care unit survivors of sepsis

Garrity, Kevin Mark (2026) Characterisation of cardiovascular function in intensive care unit survivors of sepsis. PhD thesis, University of Glasgow.

Abstract

There are over 200,000 intensive care unit (ICU) admissions in the UK each year, with sepsis being one of the most common admission diagnoses. Over two thirds of ICU survivors experience physical, cognitive or psychological impairments following discharge, often-termed post-intensive care syndrome (PICS). However, the mechanisms by which these impairments occur is unclear.

Patients admitted to ICU with sepsis often experience acute cardiovascular dysfunction, e.g. myocardial infarction, dysrhythmia, or sepsis induced cardiomyopathy. These patients often have significant haemodynamic instability such that they require infusions of vasoactive medicines to support blood pressure and other physiological targets. It is now increasingly recognised that there are often long-term cardiovascular consequences of sepsis admission. Compared to population or hospitalised controls, the incidence of stroke, heart failure and myocardial infarction are increased following an admission with sepsis in a magnitude comparable to other traditional cardiovascular risk factors, such as hypertension or dyslipidaemia. Sepsis is also known to be associated with persistent activation of inflammatory pathways that are also implicated in the pathogenesis of heart failure and other chronic cardiovascular conditions.

Thus, there are plausible mechanisms by which admission with sepsis may result in an increased risk of cardiovascular disease. There is also significant overlap between symptoms commonly experienced in patients with PICS (e.g. breathlessness and fatigue) and those experienced by patients with heart failure. The work in this thesis aims to characterise the cardiovascular function in ICU survivors of sepsis and explore whether cardiovascular dysfunction may be associated with the functional impairments seen in PICS.

Chapter 1 presents an overview of key concepts in critical illness, PICS, sepsis and heart failure. It outlines definitions of critical illness, the long-term consequences of ICU admission and the manifestations of PICS. It highlights key concepts in sepsis and the associations with chronic inflammation during convalescence. Finally, the chapter discusses cardiovascular dysfunction during sepsis and how it may be implicated in the pathogenesis of long term cardiovascular disease.

Chapter 2 presents a scoping review of the literature surrounding cardiovascular dysfunction in survivors of sepsis. It demonstrates that whilst there is an increasing body of literature that recognises the risk of adverse cardiovascular events associated with sepsis admission, there is a paucity of data that seeks to address the underlying mechanisms by which this phenomenon occurs.

Chapter 3 presents the methodology used in a multi-site observational cohort study utilising cardiovascular magnetic resonance (CMR) imaging, biomarker analysis and patient reported outcome measures (PROMs) to characterise the cardiovascular function in a cohort of ICU survivors of sepsis from Scottish ICUs.

Chapter 4 describes the overall demographics of 50 patients recruited to the ICU follow-up study after sepsis admission. Of these patients, 34 were successfully followed up with CMR imaging, biomarker analysis and PROMs.

Chapter 5 describes the CMR findings in the study cohort, 6-10 weeks following hospital discharge. Left ventricular systolic dysfunction (LVSD) was present in 7/34 (20.6%) of the study population and was associated with acute LV dysfunction during ICU admission (OR 7.67 (95%CI 1.26, 54.5) p=0.030). The LVSD was global in nature, with no evidence of late gadolinium enhancement and scar in any participants. LVSD was also associated with N terminal pro-B-type natriuretic peptide (NT-proBNP) and high sensitivity troponin (hs-TnT) at 6-10 weeks post hospital discharge. The cohort as a whole had elevated median values of native T1 and ECV, indicating fibrosis and expansion of the extracellular volume.

Chapter 6 describes cardiovascular, renal and inflammatory biomarkers in ICU survivors of sepsis at recruitment (at the time of ICU discharge) and at follow-up (6-10 weeks post hospital discharge). It demonstrates elevated levels of the cardiovascular biomarker NT-proBNP (median 1,446pg/mL) at discharge from ICU. Median values were significantly different at follow-up in participants who had LVSD (208 vs 74pg/mL, p=0.024) and were at levels that (dependent on symptom burden and imaging findings) would be supportive of a diagnosis of heart failure. Participants with LVSD also had values of inflammatory biomarkers elevated above their normal reference range, many of which have been highlighted of pathological significance in heart failure.

Chapter 7 describes patient reported outcome measures of exercise capacity and health related quality of life (HRQoL) in the study population. Median Duke activity status index (DASI) score was 21, indicating reduced exercise capacity. As measured by EuroQol 5-level 5-dimension (EQ-5D-5L), there were widespread levels of impairment across multiple domains. Whilst not statistically significant, participants with LVSD had reduced exercise capacity, increased measures of dyspnoea and reduced HRQoL measures in comparison to those without LVSD. Clinical frailty score prior to ICU admission was significantly associated with reduced log DASI score (β=-0.30 (95%CI -0.48, -0.12) p=0.002) and EQ-5D-5L visual analogue scale (β=-6.24 (95%CI -12.3, -0.18), p=0.044) in the study population, underpinning its importance in determining overall functional trajectory upon recovery.

These findings indicate that follow-up with CMR imaging, biomarkers and PROMs was feasible and well tolerated in this population, with attrition rates in keeping with other long-term outcome literature. LVSD was prevalent and associated with abnormal biomarkers of cardiovascular dysfunction and injury. Measures of exercise capacity and HRQoL were reduced, but not to a degree which was statistically significant. The findings are supportive of the hypothesis that chronic cardiovascular dysfunction is prevalent following ICU admission with sepsis, and may be associated with both the acute cardiovascular dysfunction seen during index illness and ongoing inflammation during convalescence. Furthermore, these data suggest there may be a cohort of patients at increased risk of cardiovascular disease following ICU who could benefit from established therapy to modify recovery trajectory and reduce morbidity following admission.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Additional Information: This thesis was funded via the Wellcome Trust Translational Partnership Award and the Fiona Elizabeth Agnew Trust (FEAT).
• Subjects: R Medicine > RC Internal medicine
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Medicine, Dentistry & Nursing
• Funder's Name: Wellcome Trust Translational Partnership Award, NHS Greater Glasgow and Clyde (NHSGGC)
• Supervisor's Name: McCall, Dr. Philip and McPeake, Dr. Joanne
• Date of Award: 2026
• Depositing User: Theses Team
• Unique ID: glathesis:2026-86246
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 24 Sep 2026 10:36
• Last Modified: 24 Sep 2026 10:37
• Thesis DOI: 10.5525/gla.thesis.86246
• URI: https://theses.gla.ac.uk/id/eprint/86246

Related URLs

• Article DOI
• Article DOI
• Article DOI
• Conference item
• Conference item