Dyslipidemia in the Critically Ill

Dyslipidemia in the Critically Ill
In 1926, Thannhauser and Schaber [1] reported an association between low cholesterol and disease. Subsequently, total cholesterol, low-density lipoprotein (LDL), and high-density lipoprotein (HDL) have been shown to be substantially reduced in patients with many disorders including infections, burns, and cancer [2–4]. In addition, a fall in serum cholesterol has been noted in patients who undergo surgical interventions and myocardial infarction [5–7]. Hypocholesterolemia has recently been reported in patients who have sepsis and in critically ill and injured patients [8–10]. The serum cholesterol in many of the study subjects fell below 100 mg/dL, well within the range associated with severe malnutrition
[11]. In patients who have severe sepsis, total and HDL cholesterol levels fall rapidly and reach 50% of the recovery levels by day 3, followed by a slow increase over the next 28 days [10]. The hypocholesterolemia of acute illness is usually associated with a moderate increase in triglyceride levels caused by an increase in very low-density lipoprotein (VLDL). The degree of hypocholesterolemia
appears to correlate with the severity of illness, morbidity, and mortality.
In a large cohort of hospitalized patients who had cholesterol levels b 100mg/dL, a ten-fold higher mortality was reported, which inversely correlated with cholesterol levels [12]. Gordon and colleagues [9] demonstrated that low serum cholesterol levels on admission to a surgical ICU were associated with a higher APACHE III and Multi-Organ Dysfunction Score (MODS), longer length of stay, and higher mortality. Bonville and colleagues [13] demonstrated that decreased serum cholesterol was an independent predictor of MODS and mortality. Dunham and colleagues [14] followed the total serum cholesterol in 28 ventilated trauma patients and found the cholesterol at admission was decreased compared with expected values. The serum cholesterol increased in survivors and fell in patients who died. These researchers also noted that the cholesterol level fell at the onset of infection, and that fall in cholesterol was more predictive of infection than an increase in white cell count. Similarly, Sun and colleagues [6] reported that the serum cholesterol, serum albumin, and total protein decreased by approximately
40% in a large cohort of patients who had undergone major surgery. In this study, the serum cholesterol increased significantly after nine days of enteral nutrition. Obialo and colleagues [15] reported that hypocholesterolemia (total cholesterol b 150 mg/dL) and hypoalbuminemia (serum albumin b 3.5 g/dL)
were independent predictors of mortality in patients with acute renal failure.
Hypocholesterolemia may be part of the negative acute phase response to acute illness. The mechanism for the low cholesterol and lipoprotein concentrations in acute illness is most likely multifactorial; both decreased synthesis and enhanced catabolism occur. Giovannini and colleagues [16] demonstrated that hypocholesterolemia in critically ill surgical patients correlated with the decrease
in plasma proteins and hepatic protein synthesis. Increased levels of proinflammatory cytokines may explain the hypocholesterolemia of acute illness.
Decreased synthesis of apoproteins has been demonstrated in hepatic cell lines exposed to tumor necrosis factor-a (TNF-a), interleukin-1b (IL-1b) and interleukin 6 [17]. The parenteral administration of pro-inflammatory cytokines has been demonstrated to lower lipid levels [18,19]. Furthermore, Gordon and colleagues [9] have demonstrated an inverse correlation between interleukin-6 and
apolipoprotein A-1 levels in surgical ICU patients.
An important issue is whether low lipid concentrations are of physiologic significance or whether hypolipidemia merely reflects the degree of acute illness.
Emerging data suggest that hypolipidemia may have important clinical consequences. Low lipid and lipoprotein concentrations were associated with a poor prognosis in several studies of elderly individuals [20–22] and correlated with the development of infectious disorders over a 15-year period in the Kaiser Permanente study of 15,000 healthy men and women [23]. Low total and HDL
cholesterol have been associated with an increased risk of nosocomial infections in surgical patients [24]. Chenaud and coworkers [25] demonstrated that surgical patients who had low apolipoprotein A-1 levels on ICU admission were significantly more likely to have a systemic inflammatory response syndrome exacerbation when compared with patients who had higher apolipoprotein A-1 levels. Leardi and colleagues [26] reported the risk of post-operative infections to be 73% among patients who had a total cholesterol concentration b 105 mg/dL compared with an incidence of 35% (P b 0.001) in patients who had a higher cholesterol concentration. Lipoproteins, especially HDL, bind to and neutralize
lipopolysaccharide (LPS). Reconstituted HDL (rHDL), which consists of purified apolipoprotein A-1 and phosphatidylcholine is even more effective in neutralizing endotoxin toxicity [27]. Lipoproteins may compete with LPS binding protein (LPSBP) for binding LPS. LPS bound to LPSBP activates the CD14 toll-like receptor complex on mononuclear cells, which then activates the pro-inflammatory
cascade. Animal experiments appear to corroborate this interaction. Transgenic
152 marik mice that have elevated HDL or LDL concentrations are protected against lethal
endotoxemia and severe gram-negative infections [28,29]. In these experiments, the transgenic animals have lower free levels of endotoxin, lower cytokine levels, and improved survival when compared with wild-type animals. Infusions of a reconstituted lipoprotein that contained apolipoprotein A-1 and phosphatidylcholine, blocked LPS-induced cytokine production with a significant attenuation
of hypotension, acidosis, and leukopenia in a rabbit endotoxin shock model [30]. Similarly, a reconstituted HDL formulation protected mice against lethal doses of LPS [28]. Gordon and colleagues [8] have demonstrated that LPS-stimulated production of TNF-a in blood from critically ill patients and
controls was completely suppressed by the addition of a reconstituted HDL preparation. Besides the effect of HDL and rHDL on LPS toxicity, HDL can also influence the fibrinolytic pathway and platelet function directly. Humans who have a high circulating concentration of HDL have higher tissue-type
plasminogen activator and plasminogen activator inhibitor concentrations than subjects with low HDL levels [31]. HDL affects platelet function through interaction with the glycoprotein IIb-IIIa complex, and thereby competes with the binding of fibrinogen to platelets and results in inhibition of platelet aggregation [32]. In a double-blind, placebo-controlled, crossover study, Pajkrt and
colleagues [33] injected LPS following an infusion of rHDL or placebo into healthy male volunteers. In this experiment, rHDL significantly reduced LPSinduced activation of coagulation and fibrinolysis and collagen stimulated platelet aggregation.
These data suggest that although decreased total and HDL cholesterol may reflect disease severity and be a manifestation of the negative acute phase response, the low cholesterol levels may predispose critically ill patients to endotoxemia, sepsis, and MODS. However, it is likely that the adverse sequella
of hypocholesterolemia may operate through other mechanisms. Specifically, low levels of HDL may lead to adrenal insufficiency in critically ill patients. Adrenal insufficiency is being recognized with increasing frequency in critically ill patients: a reported incidence of up to 61% in patients who experience septic shock [34,35]. The mechanisms that lead to reversible adrenal failure of the critically ill are poorly understood; however, low HDL levels may play an important role [36].
The adrenal gland does not store cortisol; increased secretion occurs because of increased synthesis controlled by adrenocorticotropin (ACTH). Cholesterol is the principal precursor for steroid biosynthesis in steroidogenic tissue. In a number of sequential enzymatic steps, cholesterol is metabolized by P450
cytochromes to aldosterone, dehydroepiandrosterone, androstenedione, and cortisol.
The first and rate-limiting step is the formation of pregnenolone from cholesterol. At rest and during stress about 80% of circulating cortisol is derived from plasma cholesterol; the remaining 20% is synthesized in situ from acetate and other precursors [37]. Experimental studies suggest that HDL is the preferred cholesterol source of steroidogenic substrate in the adrenal gland [38].
It is postulated that the decreased delivery of substrate (HDL) for cortisol synthesis at a time of increased demand may result in ‘‘relative adrenal insufficiency’’ [36]. Recent studies support this hypothesis. van der Voort and colleagues [44] performed adrenal function testing in a cohort of critically ill patients with MODS. In this study, 38% of patients had an abnormal high-dose synacthen stimulation test. The HDL levels were 28 mg/dL in the responders compared with 12.8 mg/dL in the non-responders (P = 0.02). By multivariate analysis, HDL cholesterol was the sole predictor of cortisol response. Similarly,
Marik and colleagues performed adrenal function testing in 340 patients who had liver disease. In this study, 72% of patients were diagnosed with adrenal insufficiency (hepato-adrenal syndrome). The HDL level at the time of adrenal testing was the only variable predictive of adrenal insufficiency (P b 0.0001).
These data suggest that although total and HDL cholesterol levels fall with acute illness, the low cholesterol levels may predispose critically ill patients to endotoxemia, sepsis, and adrenal failure.
_________________________________________________________________________________
Part of the article. Whole article and references:
Crit Care Clin 22 (2006) 151– 159
Paul E. Marik, MD
Division of Pulmonary and Critical Care Medicine, Thomas Jefferson University,
1015 Chestnut Street, Suite M-100, Philadelphia, PA, 19107, USA
Capitulo Bioquimico - Sociedad Argentina de Terapia Intensiva
http://www.sati.org.ar