BALTIMORE—Sleep-disordered breathing (SDB) and insulin resistance are each associated with obesity, but a new study links SDB with insulin resistance even among non-obese subjects.[1] A separate investigation of mildly obese men supports an independent association between SDB and insulin resistance: After accounting for the effects of body-mass index (BMI) and percent body fat, researchers found that men who suffered five or more apneic/hypopneic events per hour were more than twice as likely to have impaired glucose tolerance than were non-apneic subjects.[2] Further, both studies found a direct association between glucose intolerance and degree of oxygen desaturation during sleep.

Patients with obstructive sleep apnea (OSA) frequently present with a cluster of disorders, including central obesity, hyperinsulinemia, glucose intolerance, dyslipidemia, and hypertension. This syndrome, which may predispose a patient to cardiovascular risk, is thought to be produced by insulin resistance and the resultant decrease in insulin-mediated glucose uptake.

“Clearly, obesity is associated with sleep-disordered breathing, and obesity is associated with insulin resistance,” said Naresh M. Punjabi, MD, PhD, lead author of the second study. But the question remains: Is the association between insulin resistance and SDB secondary to their respective linkage with obesity? Using a sample of healthy but overweight men, Dr. Punjabi et al attempted to determine whether sleep apnea is independently associated with glucose intolerance.

AHI, SAO2 LINKED TO RESISTANCE

Ip and colleagues investigated OSA’s impact on risk for insulin resistance in 270 patients (73 female) who had no prior diagnosis of diabetes.[1] All patients had been referred for and underwent polysomnography, and each patient’s fasting glucose (G0) and fasting insulin (I0) were also measured. Of these patients, 185 (69%) were found to have an apnea-hypopnea index (AHI) of five events or more per hour and thus were given a diagnosis of OSA. As compared with non-apneic subjects, OSA patients were older and more obese; they also had higher I0 levels, as well as greater values on the homeostasis model assessment (HOMA; G0I0/22.5), a measure of insulin resistance.

Stepwise linear regression revealed obesity as the strongest factor predisposing patients to insulin resistance. However, AHI was also an independent predictor, with each additional apneic or hypopneic event per hour approximately corresponding to a 0.5% increase in I0 level and HOMA. Minimum arterial oxygen saturation (Sao2), another physiologic parameter of OSA severity, also independently predicted I0 and HOMA.

The relationship between OSA and insulin resistance held among obese and non-obese patients alike. Furthermore, insulin resistance was associated with hypertension in this population: Hypertensive subjects had significantly higher I0 and HOMA values.

CORROBORATING EVIDENCE

These findings were largely corroborated by the results of Dr. Punjabi and colleagues, which were published simultaneously. Punjabi et al examined 150 healthy men 45 years or older who weighed 120% to 160% of their respective ideal body weights. Polysomnography revealed sleep apnea (defined by an AHI of five or more events per hour) in 62% of these participants. Confirming previous findings, this study showed that AHI was correlated with BMI. Additionally, two-hour glucose levels increased with AHI, and elevated insulin resistance was associated with increasing AHI.

After adjusting for BMI and percent body fat, however, Dr. Punjabi and coworkers found that sleep apnea was independently associated with impaired glucose tolerance, with an odds ratio (OR) of 2.15. Independent of BMI and body fat, insulin resistance was also associated with SDB: For every 5% increase in AHI, G0/I0 increased by 0.25, and HOMA increased by 2.11.

Examining individual aspects of SDB that might contribute to deficits in glucose metabolism, the researchers then assessed the effects of apnea-related oxyhemoglobin desaturation on glucose metabolism. After adjustment for body fat, BMI, and AHI, severity of glucose intolerance paralleled the extent of oxygen desaturation: For every 4% decrease in mean minimum Sao2 during apneic/hypopneic events, risk for glucose intolerance nearly doubled (OR, 1.99). Likewise, a 2% decrease in nightly minimum Sao2 was independently linked with a drop in G0/I0 of 0.32, indicating an independent association with insulin sensitivity as well.

DOES SDB CAUSE INSULIN RESISTANCE?

SDB’s association with glucose intolerance suggests a causal relationship and thus hints that measures to counter SDB could prove useful in preventing and combating diabetes in patients at risk. Unfortunately, the evidence for an effect of interventions such as continuous positive airway pressure on insulin resistance is conflicting.

If SDB indeed has a causal role, it remains unclear which aspects of SDB might contribute to changes in glucose metabolism—or how. Dr. Punjabi, an Assistant Professor of Medicine and Epidemiology at Johns Hopkins University, suggested two features of SDB that may be important: oxygen desaturation and frequent awakening. In favor of the former, he cited evidence from a rat model of SDB demonstrating that hypoxia can actually induce insulin resistance.[3,4]

On the other hand, Dr. Punjabi pointed out, repeated awakening could also take its toll. “There are some data that suggest that … insufficient sleep is related to glucose intolerance and insulin resistance.” He described a study of healthy young adults with normal glucose metabolism whose sleep was limited to four hours a night for six days.[5] Glucose tolerance in these subjects was lowered following sleep deprivation. Dr. Punjabi extrapolated this finding to his own work: “One can take that one step further and say that since patients with sleep apnea have a state of secondary chronic insufficient sleep from recurrent awakenings, they may also be predisposed to altered metabolism.”

Decreased glucose tolerance in sleep-deprived subjects was accompanied by heightened sympathetic activation and elevated cortisol levels, suggesting possible involvement of these factors in a causal mechanism. “It’s well known that sympathetic activation leads to glucose mobilization.” Thus, Dr. Punjabi explained, “a continuously adrenergically driven system would have a high likelihood of developing glucose intolerance.” SDB patients might share such an effect with intentionally sleep-deprived subjects, he reasoned: “There is clear evidence that apneics have active sympathetic systems,” a finding also cited in proposed mechanisms for elevated cardiovascular risk.

Whether oxygen desaturation or sleep deprivation per se plays a greater role in elevating SDB patients’ risk for glucose intolerance remains open to debate, in Dr. Punjabi’s opinion. “Is it just hypoxemia, ... or is it arousal? I can’t say, but there are two schools of thought,” he noted. “It’s certainly possible that … both can contribute independently to the insulin resistance.” While he and his colleagues did not have detailed data on subjects’ arousal during the study, they now plan to examine the associations between awakenings and desaturation.

—Mimi Zucker, PhD

References
1. Ip MSM, Lam B, Ng MMT, et al. Obstructive sleep apnea is independently associated with insulin resistance. Am J Respir Crit Care Med. 2002;165:670-676.
2. Punjabi NM, Sorkin JD, Katzel LI, et al. Sleep-disordered breathing and insulin resistance in middle-aged and overweight men. Am J Respir Crit Care Med. 2002;165:677-682.
3. Raff H, Bruder ED, Jankowski BM, et al. Effect of neonatal hypoxia on leptin, insulin, growth hormone and body composition in the rat. Horm Metab Res. 2001;33:151-155.
4. Raff H, Bruder ED, Jankowski BM. The effect of hypoxia on plasma leptin and insulin in newborn and juvenile rats. Endocrine. 1999;11:37-39.
5. Spiegel K, Leproult R, Van Cauter E. Impact of sleep debt on metabolic and endocrine function. Lancet. 1999;354:1435-1439.

Return to table of contents