Breathing becomes shallower and lighter when falling asleep

Key Points

• CO2 increases significantly prior to or simultaneously with sleep onset

• Breathing becomes shallower with sleep onset

• Breathing rate does not change with sleep onset

The Breathing Diabetic Summary

We have reviewed several studies on sleep and breathing (here, here, and here).  However, we have not looked at how breathing changes as we fall asleep.  This study sheds light on central nervous system changes occurring during sleep onset.

Four women and 8 men that had no history of respiratory or sleep disorders were studied. They were instructed to relax in a reclined bed in the dark and go to sleep for ~1 hour.  Five of the subjects also were asked to relax in the bed, but stay awake for 45 minutes for control measurements.

(This is a very small sample size, so let’s remember that as we go through the results.)

Measurements of alveolar CO2 and chest and abdominal breathing motions were taken to reflect central nervous system changes during sleep onset.  Breathing rate also was measured.  The subjects went through 3 separate sessions to acclimate to the laboratory setting.

During all three sessions, when the participants fell asleep, their CO2 rose significantly and their breathing became shallower.  The significant increase in CO2 was not associated with a change in breathing rate.  This implies that breathing became not only shallower, but also lighter.

In every case, CO2 rose and breathing became shallower either prior to or simultaneously with falling asleep.  And because breathing rate did not change, the rise in CO2 indicates that the subjects were breathing less.

The close correlation between CO2, shallow breathing, and sleep onset indicates that breathing patterns might be a reliable indicator of drowsiness in a person.

Abstract

This study provides a systematic examination of factors that may contribute to respiratory changes associated with sleep onset. The electroencephalogram, alveolar CO2 tension, patterns of abdominal and thoracic respiratory movements, and respiratory rate were measured in three sessions each on 12 normal subjects as they fell asleep, and also on 5 of them as they lay awake. Nonintrusive respiration measurement devices were used. Resting awake CO2 tension was found to increase significantly across sessions. In addition, CO2 tension was significantly higher during stages 1 and 2 of sleep than during wakefulness on days 2 and 3. There was also a shift from relatively greater abdominal expansion toward relatively greater thoracic expansion with sleep onset. None of these changes occurred when subjects remained awake during a session. We conclude that changes in respiration with sleep onset cannot be accounted for solely by changes due to habituation, merely lying quietly, or the effects of the measuring devices. Rather, they appear to be caused by a central interaction between centers controlling the level of wakefulness and those controlling respiration.

Journal Reference:

Naifeh KH, Kamiya J.  The nature of respiratory changes associated with sleep onset.  Sleep.  1981;4(1):49-59.

Treat & reverse the root cause of diabetic complications (tissue hypoxia) with slow breathing

Key Points

• Type-1 diabetics exhibit lower resting oxygen saturation, lower cardiovascular control, reduced hypoxic chemoreflexes, and enhanced hypercapnic chemoreflexes

• The root cause of these problems is resting tissue hypoxia, which causes over-activation of the sympathetic nervous system and autonomic and cardiovascular dysfunction

• Autonomic imbalance in diabetes is largely functional, and therefore reversible

The Breathing Diabetic Summary

This is a follow-on to our previous paper on cardio-respiratory control in diabetes.  This paper, however, is a clinical study rather than a literature review.

Previous studies have shown respiratory problems in diabetics.  Previous studies also have shown cardiovascular dysfunction in diabetics.  However, no studies simultaneously examined both of these factors in an integrated fashion.  Thus, the aim of this study was to comprehensively examine cardio-respiratory function in type-1 diabetics.

The key measurements from this paper were resting oxygen saturation, baroreflex sensitivity (BRS; a marker of cardiovascular and autonomic control), and both hypoxic and hypercapnic chemoreflexes (markers of respiratory control). 

Their hypothesis: If the BRS and chemoreflexes were suppressed in diabetics, this would indicate nerve damage was present.  However, if cardiovascular function was suppressed, while chemoreflexes were enhanced, this would indicate autonomic imbalance that has a functional cause.  In this latter case, therapies aimed at restoring cardio-respiratory control (for example, slow breathing) could help prevent diabetic complications.

The study had 46 patients with type-1 diabetes and 103 age-matched control subjects.  The participants went through a variety of tests to evaluate baroreflex functioning and chemoreflexes.  For example, to measure the patients’ hypercapnic chemoreflex, oxygen was kept constant while CO2 was gradually increased.  The chemoreflex can then be measured as the slope of the relationship between minute ventilation and change in CO2 (or oxygen in the case of the hypoxic chemoreflex).  A large change in minute ventilation for a small change in CO2 would represent an enhanced hypercapnic chemoreflex.

Interestingly, the results showed that although diabetics displayed larger breathing volumes than controls, they had slightly higher CO2 levels and reduced oxygen saturation.  However, they did have an enhanced hypercapnic chemoreflex, meaning they could not tolerate changes in CO2 as well as controls.  And, somewhat surprisingly, they had a reduced hypoxic chemoreflex, meaning they could tolerate lower oxygen levels without increasing their breathing as much as controls.

The diabetics also exhibited a lower resting oxygen saturation. This is fascinating because the lower resting oxygen saturation implies a significantly reduced partial pressure of oxygen (due to the oxyhemoglobin dissociation curve). This would result in tissue hypoxia. What’s more, they cite a paper (which is now near the top of my reading list) that shows that a high HbA1c also reduces tissue oxygenation by increasing oxygen’s affinity to hemoglobin (shifting the dissociation curve to the left). 

The authors suggest that their results can be interpreted as follows: Resting tissue hypoxia, combined with a suppressed hypoxic chemoreflex, leads to an enhanced compensatory hypercapnic chemoreflex and chronic activation of the sympathetic nervous system.  This, in turn, leads to a suppression of the cardiovascular system (reduced BRS and reduced heart rate variability).  It’s a vicious cycle.

However, this is actually great news.  Their results suggest that diabetic autonomic imbalance is largely functional and not related to nerve damage.  (Remember, both the cardiovascular reflexes and the chemoreflexes would have been suppressed with nerve damage).  In fact, the authors suggest that this imbalance likely leads to nerve damage rather than being the result of it. Therefore, therapies targeting cardio-respiratory control could help reverse/prevent diabetic complications.

Finally, the authors suggest that breathing control and physical exercise could be two such therapies to restore cardio-respiratory function.  We know that slow breathing has many therapeutic benefits for the cardiovascular, autonomic, and respiratory systems.  And, we know that slow, light breathing increases CO2 and increases tissue oxygenation (due to the Bohr effect).  Now, we know that these positive benefits have the potential to stop or reverse diabetic complications. 

Abstract from Paper

BACKGROUND: Cardiovascular (baroreflex) and respiratory (chemoreflex) control mechanisms were studied separately in diabetes, but their reciprocal interaction (well known for diseases like heart failure) had never been comprehensively assessed. We hypothesized that prevalent autonomic neuropathy would depress both reflexes, whereas prevalent autonomic imbalance through sympathetic activation would depress the baroreflex but enhance the chemoreflexes.

METHODS: In 46 type-1 diabetic subjects (7.0±0.9year duration) and 103 age-matched controls we measured the baroreflex (average of 7 methods), and the chemoreflexes, (hypercapnic: ventilation/carbon dioxide slope during hyperoxic progressive hypercapnia; hypoxic: ventilation/oxygen saturation slope during normocapnic progressive hypoxia). Autonomic dysfunction was evaluated by cardiovascular reflex tests.

RESULTS: Resting oxygen saturation and baroreflex sensitivity were reduced in the diabetic group, whereas the hypercapnic chemoreflex was significantly increased in the entire diabetic group. Despite lower oxygen saturation the hypoxic chemoreflex showed a trend toward a depression in the diabetic group.

CONCLUSION: Cardio-respiratory control imbalance is a common finding in early type 1 diabetes. A reduced sensitivity to hypoxia seems a primary factor leading to reflex sympathetic activation (enhanced hypercapnic chemoreflex and baroreflex depression), hence suggesting a functional origin of cardio-respiratory control imbalance in initial diabetes.

Journal Reference:

Bianchi L, Porta C, Rinaldi A, Gazzaruso C, Fratino P, DeCata P, Protti P, Paltro R, Bernardi L. Integrated cardiovascular/respiratory control in type 1 diabetes evidences functional imbalance: Possible role of hypoxia. Int J Cardiol. 2017;244:254 – 259.