An investigation of the measurements of vitamin status and outcome in patients with critical illness

Vasilaki, Aikaterini (2010) An investigation of the measurements of vitamin status and outcome in patients with critical illness. PhD thesis, University of Glasgow.

Abstract

This thesis describes a prospective observational longitudinal study that examines the relationship between the systemic inflammatory response, vitamin concentrations and outcome in patients with critical illness. Venous blood samples were collected from patients admitted in the Intensive Care Unit of Glasgow Royal Infirmary on admission (n=126) and on follow-up (n=77). The concentrations of vitamins B2, B6, C, A, E, lutein, lycopene, α-carotene and β-carotene in plasma, red cells and white cells and plasma total and free malondialdehyde, that was used as a marker of lipid peroxidation, were assessed by high performance liquid chromatography (HPLC).
In Chapter 3, vitamin B2 was examined. The results of this study showed that the ratio of plasma FAD to riboflavin in critically ill patients on admission and on follow-up was much lower than that of the plasma ratio in the controls. There were also similar findings with respect to the red cell FAD to riboflavin ratios in controls and critically ill patients. However, on admission the reduction of the FAD to riboflavin ratio was greater in the plasma (83%) compared with the red cells (49%). These results showed perturbation of the relationship between plasma FAD and riboflavin in patients with critical illness. Moreover, this appeared to be primarily due to a relative reduction in plasma and intracellular FAD concentrations although red cell FAD concentrations were all within the reference interval.
In Chapter 4, vitamin B6 was examined. The results of this study showed that, in plasma, red cells and white cells, PLP is strongly and directly associated with the concentrations of PL in patients with critical illness. The ratio of plasma PLP to PL in critically ill patients on admission and on follow-up was much lower than that of the plasma controls. There were also similar findings with respect to the red cell PLP to PL ratios in controls and critically ill patients. However, on admission the reduction of the PLP to PL ratio was greater in the plasma (55%) compared with the red cells (18%). These results report that the relationship between plasma PLP and PL in controls is similar to that found in red and white cells, in patients with critical illness. Given that, compared to plasma PLP, concentrations of plasma PL were more strongly correlated with those in the red or white cells and that red cell and white cell PL concentrations were strongly and similarly correlated with their respective PLP concentrations, demonstrates the importance of PL in the intracellular metabolism of PLP in normal subjects and critically ill patients. One interpretation of these results could suggest that plasma PL may be a good surrogate measure of intracellular PLP concentrations. However, PL is not the physiologically active form of vitamin B6 and therefore, the clinical relevance of measuring plasma PL concentrations is not clear. Therefore, the present intracellular measurements of PLP, compared with plasma, may be a more accurate reflection of vitamin B6 status.
In Chapter 5, vitamin C was examined. The results of this study showed that plasma ascorbic acid concentrations in critically ill patients were below reference intervals whereas white cell ascorbic acid concentrations were within the reference intervals. Moreover, plasma and intracellular concentrations of both vitamin C and α-tocopherol were poorly correlated and did not increase with supplementation in the ICU. Taken together these results would suggest that plasma ascorbic acid concentrations may poorly reflect intracellular concentrations in patients with critical illness.
In Chapter 6, vitamin E was examined. The results of this study showed that, compared with control subjects, the critically-ill patient group had lower plasma α-tocopherol concentrations and also when expressed per mmol of triglycerides. In contrast, α-tocopherol concentrations were higher when expressed per mmol of cholesterol despite these patients receiving no vitamin E supplementation in ICU. Moreover, neither plasma α-tocopherol nor plasma α-tocopherol expressed per mmol of lipids was strongly correlated with red cell α-tocopherol concentrations in the critically-ill patients. Finally, median red cell α-tocopherol concentrations were the same in the healthy subjects and patients with critical illness and systemic inflammatory response syndrome. Taken together the results indicate that plasma α-tocopherol alone or expressed per mmol of lipids (using either cholesterol or triglycerides) may be a less reliable marker of vitamin E status than red cell α-tocopherol in patients with a systemic inflammatory response.
In Chapter 7, lipid soluble antioxidants A, E and carotenoids and lipid peroxidation, as measured by malondialdehyde (MDA, total and free), were examined. The results of this study showed that, on admission to ICU, the free MDA fraction was increased and carotenoid concentrations were low in patients with critical illness. The majority of patients had plasma retinol concentrations below the reference interval on admission to ICU and these were directly and significantly related to α-tocopherol and the carotenoids. Moreover, this relationship was maintained on follow-up. Both albumin and C-reactive protein were consistently correlated with these lipid soluble antioxidant concentrations. Therefore, the reduction of lipid soluble antioxidant concentrations, in particular carotenoids, is likely to be multifactorial and not solely dependent on consumption during lipid peroxidation.
In Chapter 8, the behaviour of plasma and intracellular B2, B6 and C vitamin concentrations during supplementation in patients with critical illness was studied. The results of the present study show a discrepancy in the behaviour of plasma and intracellular B2, B6 and C vitamin concentrations during supplementation in patients with critical illness. Indeed, on admission to ICU, almost 1/3 of the patients had plasma vitamin B2, B6 and C concentrations below the reference interval. In contrast, on admission to ICU, red cell FAD, red cell PLP and white cell ascorbic acid concentrations were significantly higher in the patients with records of prior supplementation compared with the patients without. The longitudinal findings were in agreement with this observation and taken together, this study may suggest that cellular concentrations of vitamins B2, B6 and C may be more accurate markers of their respective vitamin status and may be expected to be a more true reflection of the presence of supplementation than their respective plasma concentrations.
In Chapter 9, the relationship between lipid peroxidation, water and lipid soluble vitamins and severity of illness or hospital outcome in critically ill patients was examined. The results of this study showed that lipid peroxidation, as evidenced by an elevated free MDA fraction, was most significantly associated with death in hospital. The concentrations of retinol and the carotenoids in the plasma were well below the reference intervals and although correlated with the free MDA fraction, had no independent prognostic value. Albumin was directly associated with the lipid soluble vitamins in the plasma, in particular retinol, α-tocopherol, lutein and lycopene. Albumin was shown to have prognostic value independent of APACHE II and when albumin was removed from the multivariate model, lycopene became an independent predictor of hospital death. None of the intracellular vitamin concentrations were related to severity of illness in critically ill patients. Taken together these results would suggest that low plasma levels of lipid soluble vitamins may be due to redistribution in the same way that albumin concentrations fall as part of the systemic inflammatory response syndrome.
In summary, the current thesis showed that, during the systemic inflammatory response, intracellular vitamin concentrations may be more accurate than plasma concentrations as indicators of vitamin status in patients with critical illness.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Keywords: critical illness, nutrition, vitamins, red cells, white cells, plasma, outcome, mortality
• Subjects: R Medicine > R Medicine (General); R Medicine > RZ Other systems of medicine
• Colleges/Schools: College of Medical Veterinary and Life Sciences
• Supervisor's Name: Kinsella, Professor John, McMillan, Professor Donald C and Talwar, Dr. Dinesh
• Date of Award: 2010
• Depositing User: Miss Aikaterini Vasilaki
• Unique ID: glathesis:2010-1829
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 24 May 2010
• Last Modified: 10 Dec 2012 13:47
• URI: https://theses.gla.ac.uk/id/eprint/1829