© 2004 American Thoracic Society
Improvement of Metabolic Function in Sleep ApneaThe Power of Positive PressureJohns Hopkins University Baltimore, Maryland Sleep apnea is chronic condition characterized by upper airway collapse during sleep. The resulting decrease or cessation of airflow is often associated with recurrent drops in the oxyhemoglobin saturation and sleep fragmentation. There is little debate that sleep apnea is causally associated with neurobehavioral consequences including daytime sleepiness and impaired quality of life. Over the last two decades there has been considerable research on whether sleep apnea is also a risk factor for adverse cardiovascular outcomes. Population-based data, to date, suggest that sleep apnea is independently associated with incident hypertension (1) and prevalent cardiovascular disease (2). Although the intermediate mechanisms between sleep apnea and cardiovascular risk are not well defined, it is generally accepted that intermittent hypoxia and sleep fragmentation play an important role in the putative causal pathway. Because metabolic dysfunction is associated with increased cardiovascular risk, a number of groups have also examined the metabolic consequences of sleep apnea. Experimental data from animal (3, 4) and human studies (5, 6) indicate that exposure to hypoxia can perturb glucose homeostasis. Moreover, curtailment of sleep duration with ensuing sleep loss can also have detrimental effects on glucose regulation (7). Thus, individuals with sleep apnea may be at risk for metabolic dysfunction due to the effects of intermittent hypoxia and/or sleep disruption. If sleep apnea provides the necessary physiologic milieu for metabolic dysfunction, what is the strength of evidence to indicate that an association, and in particular a causal association, exists between the two disorders? In a recent systematic review (8), it was noted that most, but not all, of the available studies provide strong support for a cross-sectional association between sleep apnea and metabolic dysfunction that is independent of the confounding effects of obesity. The presence of an independent relationship between sleep apnea and metabolic dysfunction, while necessary, is by itself insufficient for causal inference. Definitive evidence from clinic or population-based samples in which sleep apnea is an antecedent to metabolic dysfunction is lacking. It is certainly possible that despite rigorous "adjustments" for covariates, residual confounding due to obesity or other omitted variables (e.g., regional adiposity) may explain the observed co-relation. Experimental evidence, such as that derived from studies that use positive-pressure therapy to obviate the occurrence of apneas and hypopneas, provides an alternative method for addressing the question of whether sleep apnea causes metabolic dysfunction. In the current issue of the Journal (pp. 156–162), Harsch and coworkers (9) present data on the effect of positive-pressure therapy on insulin sensitivity in patients with sleep apnea. Insulin sensitivity was assessed with the hyperinsulinemic euglycemic clamp at baseline, 2 days, and 3 months after initiating therapy. Using a sample of 40 sleep apnea patients, the authors show that insulin sensitivity, as assessed by the clamp, increased by 1.04 µmol/kg · minute after 2 days and 1.79 µmol/kg · minute after 3 months of therapy. The improvement in insulin sensitivity occurred without any change in body mass index and was greater in nonobese than obese patients. Furthermore, insulin sensitivity in nonobese patients improved within 2 days, whereas no improvement was noted in the obese group until the 3-month follow-up. Based on the improvement in insulin sensitivity with positive pressure, the authors conclude that sleep apnea is an independent and causal risk factor for metabolic dysfunction. There are a number of major strengths of this study that are worth mentioning. These include the use of a "gold standard" test to measure insulin sensitivity, characterizing the time course of change, assessment of compliance with therapy, and the relatively large sample size compared with previous studies. The authors are to be commended for their work given the significant burden of performing the hyperinsulinemic clamp on a large sample of patients at different three time points. Although it is tempting to incorporate the new findings in our clinical armamentarium to improve compliance with sleep apnea therapy, the therapeutic merits of positive-pressure therapy with regard to metabolic function are not known. There are a number of issues in the study by Harsch and workers that are worth considering before a definitive conclusion on causal inference linking sleep apnea and metabolic dysfunction. First, the hyperinsulemic euglycemic clamp procedure is known to increase sympathetic neural traffic as it entails placement of intravenous catheters, frequent blood draws, and prolonged immobilization (10). Differences between baseline and follow-up measurements could be, in part, due to the fact that the patients may have acclimatized to the conditions surrounding the test during the follow-up visits. Second, the study sample at baseline was highly insulin resistant (insulin sensitivity index, 5.75 µmol/kg · minute). Normal insulin sensitivity, as measured with the clamp technique, usually exceeds 26 µmol/kg · minute at an insulin infusion rate of 1 mU/kg · minute (11). Thus, the effect of positive-pressure therapy in sleep apnea patients that are less insulin resistant than the reported patient sample needs to be determined. Finally, the clinical impact of the change in insulin sensitivity with positive-pressure therapy is not known. For comparison purposes, a 4-month strength-training program has been shown to increase insulin sensitivity from 13.5 to 16.7 µmol/kg · minute in middle-aged healthy men (12). It is certainly possible that, with continued use of positive pressure, insulin sensitivity improves further, particularly in those patients with sleep apnea and milder degrees of insulin resistance. Where does the report by Harsch and coworkers leave us? Clearly, the research on the metabolic impact of sleep apnea is in its early stages. Further studies are needed to determine the reproducibility of a treatment effect, define responsive and nonresponsive subgroups, determine the underlying mechanisms responsible for the improvement, quantify the long-term benefits of improved metabolic function on cardiovascular outcomes, and assess the impact of compliance. It is only with the availability of such studies that the question, "Does sleep apnea cause metabolic dysfunction?" can be definitively answered. For those of us seeking a better understanding of the potential link between sleep apnea and metabolic function, the report by Harsch and coworkers only marks the beginning of a challenging journey. FOOTNOTES Conflict of Interest Statement: N.M.P. has no declared conflict of interest. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1-adrenergic blockade reduces insulin sensitivity. J Appl Physiol 2001;91:623–631.