On a pathophysiological basis, fat oxidation produces less carbon dioxide (CO₂) per amount of oxygen consumed compared to carbohydrates (CHO) (38), resulting in reduced ventilatory requirements. Therefore, ketogenic diet decreases metabolic carbon dioxide production for a given oxygen consumption, and may theoretically lead to a reduction in arterial carbon dioxide partial pressure (PETCO₂), pulmonary ventilation, and CO₂ body stores (29, 30).

Carbohydrates loads have been reported to precipitate respiratory failure in patients with lung compromise (39–42) since they may worse respiratory acidosis in patients unable to improve ventilation as a compensatory mechanism to excrete more CO₂ (39, 43).

Over time, some authors investigated the effects of low-carbohydrate high-fat diets or supplements on respiratory parameters in healthy subjects, and most of them reported beneficial effects (see Table 1). In the majority of studies we report below, patients were administered an amount of CHO that could possibly lead to ketosis, but the development of ketosis wasn't verified.

Rubini et al. (29) observed that a ketogenic diet (CHO < 30 g/day) administered for 20 days reduced PETCO₂ in 32 healthy subjects, without modifications in oxygen uptake, carbon dioxide production nor expired total ventilation, which may be related to a reduction in CO₂ body stores. Therefore, it may be beneficial for patients with high carbon dioxide arterial partial pressure due to respiratory insufficiency, because it lowers CO₂ levels without increasing respiratory muscle fatigue, with the consequent risk of respiratory failure on mechanical basis. Interestingly, the reduction of PETCO₂ was maintained even after the end of the diets, suggesting a long-term effect.

Similarly, Kwan et al. (30) observed that a ketogenic diet (50 g CHO/day) administered for 1 week to 6 healthy female subjects reduced arterial carbon dioxide tension, while resting ventilation and breathing frequency remained unchanged. Interestingly, pulmonary function tests showed an increase in peak expiratory flow rate and functional residual capacity respect to the baseline, respectively, by 5 and 10%.

Sue et al. (44) studied the effects of altering the composition of dietary fat and carbohydrate content on gas exchange, at rest and during exercise, on 8 healthy volunteers. At rest, the mean oxygen uptake did not differ, as did not the minute ventilation, while mean CO₂ output was significantly less on the low-carbohydrate high-fat diet (10% of calories from CHO), compared to the high-carbohydrate diet (70% of calories from CHO). This differences was smaller during exercise, probably because of the preferential use of glycogen stores from the muscle (44).

On the other hand, one study reported negative effects of a ketogenic diet. The administration of a low-carbohydrate high-fat ketogenic diet, providing 2,400 Kcal/day for 4 weeks in 17 healthy women (<25 g of CHO/day), was associated with an earlier muscle fatigue at higher intensities and during daily life activities, probably because of the reduced glycogen stores in the muscle (45). However, a study on 7 well-trained male cyclists consuming a low-carbohydrate high-fat diet (15–82 g of CHO/day), for a long-time (at least the previous 8 months), showed that these athletes had a similar gluconeogenesis rate and reduced glycogenolysis respect to the 7 athletes fed with a mixed diet (46). These findings suggest that after a long-term adaptation to a low-carbohydrate high-fat diet, liver glycogen contributes to endogenous glucose production during exercise, and that glucose may be preferentially obtained from glycerol derived from lipolysis of intramuscular triglycerides (47), configuring an hypothetical difference among about subjects accustomed to a low-carbohydrate high-fat diet and “newbies.”

Many Chronic Obstructive Pulmonary Disease (COPD) patients experience hypercapnia and hypoxemia, therefore a nutritional approach that decreases carbon dioxide production, and consequently respiratory muscles work, is extremely interesting.

Some authors observed an improvement of ventilatory measurements in COPD hypercapnic patients after a low-carbohydrate high-fat diet. In a small sample of 14 patients with chronic hypercapnia and COPD the administration of a low-carbohydrate diet (28% CHO) for 5 days, determined a trend of lower CO₂ production, lower arterial pCO₂, respect to moderate-carbohydrate moderate-fat diet and high-carbohydrate low-fat diet. An amelioration of forced vital capacity and forced expiratory volume in 1 s (FEV₁) were observed for all diets, which may reflect an increase in muscle strength (31). However, some recent papers observed that a KD did not improve muscle strength in healthy women (45) and in trained athletes (48, 49), so we may hypothesize that the amelioration in FEV₁, which was observed for all diets, was to be ascribed to the beneficial effects of the nutritional support rather than its composition. Also, we may speculate that the effect of the diet on muscle strength would be more evident in non-trained patients than in trained athletes.

Some authors studied the short-term alterations following the administration of a particular meal or supplement. Akrabawi et al. (38) enrolled 36 outpatients with COPD and administered a high-fat meal (55% calories from fat, 25 g of CHO) or a moderate fat meal (41% calories from fat, 35 g of CHO) and observed a higher CO₂ production and O₂ consumption in patients fed the moderate-fat meal in the early post-prandial time, probably reflecting the earlier absorption of the meal, and no difference for pulmonary function. A recent study on 60 low-body weight patients with COPD and elevated arterial carbon dioxide tension observed that the group administered a low-carbohydrate high-fat evening supplement with 50–75 g of CHO daily for 3 weeks had an overall improvement in ventilatory status, with decrease in VCO₂, PaCO₂, VO₂ and minute ventilation, and an increase of PaO₂ and FEV₁, respect to the group administered a traditional high-carbohydrate diet (60–70% CHO) (50). This confirms the importance of nutritional therapy in malnourished COPD patients, since it has a strong impact on respiratory muscle weakness, and the importance of its macronutrient composition.

Many studies on the effects of carbohydrate loads on respiratory parameters and physical performance in chronic obstructive lung disease patients report a detrimental effect of CHO. High-carbohydrate diets may result in increased CO₂ production and O₂ consumption in clinically stable COPD patients (39). Kuo et al. (39) studied 12 clinically stable COPD patients and 12 healthy controls after administering an isocaloric high-fat (55.2% fat and 28.1% CHO) or high-carbohydrate (31.5% fat and 54.5% CHO) liquid meal, and found that the high-fat diet had a small effect on gas exchange parameters and ventilation, while the high-carbohydrates diet resulted in a great increase of CO₂ production, O₂ consumption and minute ventilation in COPD patients. Efthimiou et al. (40) studied a small sample of 10 clinically stable patients with the 6 min walking test. They observed that the group administered CHO-rich drink experienced a reduced physical performance correlated to the increased CO₂ production and a perceived effort to breathe. Similar detrimental effects of CHO on walking and exercise performance were obtained in 18 patients with chronic airflow obstruction (41, 42). Increased CO₂ production and increased minute ventilation after a high-carbohydrate formula, administered for nighttime enteral feeding, were observed in 10 young adult patients with cystic fibrosis with moderate to advanced lung disease (51). On the contrary, a small study with 13 patients affected by stable airways disease fed a high-carbohydrate meal (480 ml of grape juice and three-fourths cup of sucrose) showed that 7 patients who retained carbon dioxide had an increase in PaO₂, probably reflecting the increased alveolar ventilation, and had not a significant increase in PaCO₂, therefore the authors conclude that most patients with chronic airways disease are able to tolerate the increased endogenous carbon dioxide load resulting from a meal high in carbohydrates (52).

Most patients with COPD and acute respiratory failure have marked reduction of body protein stores (53) and lower muscle concentrations of adenosine triphosphate and creatinine phosphate, and these factors may be an important determinant of respiratory failure (54). Moreover, critically ill patients are more responsive to changes in dietary composition than less critical ones (55).

Some authors suggest that a low-carbohydrate diet may be an effective tool to ameliorate respiratory failure (30, 32, 33).

A study on 8 clinically stable chronic hypercapnic respiratory failure patients with congestive heart failure showed that a very low-calorie ketogenic diet (600 Kcal/day, 30 g of CHO/day), administered for on average 1 month, was effective in ameliorating arterial oxygen tension and oxygen saturation, reducing arterial carbon dioxide tension, and ameliorating electrocardiographic abnormalities associated with hypoxiemia (32). These results were probably to be ascribed to the combined effect of diet composition and weight loss, and even two non-obese patients had beneficial effects from the diet.

Another study on 8 chronic hypercapnic respiratory failure patients administered 200 or 50 g of CHO daily for a week within an isocaloric diet, observed that the reduction in the CHO intake on both diets improved the general well-being of the patients, increased arterial oxygen tension and decreased arterial carbon dioxide tension. The 50 g CHO diet compared to the 200 g CHO diet further reduced the arterial carbon dioxide tension, suggesting that such a diet may be used in patients with intractable respiratory failure (33).

Regarding mechanically ventilated patients, in the literature there are some evidences of a beneficial effect of a low-carbohydrate high-fat diet (55–57). A study on 20 clinically stable ventilated patients showed that low-carbohydrate high-fat enteral feeding (55.2% fat, 28.1% carbohydrate) is able to reduce PaCO₂ levels and the time of mechanical ventilation (55). A recent study on 100 patients with type II respiratory failure, secondary to pulmonary disease requiring mechanical ventilation, showed a great improvement in arterial carbon dioxide tension and minute volume at weaning with the low-carbohydrate high-fat feeding (55.2% fat, 28.1% carbohydrate) (56). Interestingly, similarly to Al-Saady (55), this was highly significantly associated with less time spent on mechanical ventilation (56). On the other hand, a study on 32 patients requiring mechanical ventilation reported no beneficial effects of a similar low-carbohydrate high-fat enteral nutrition (55.2% fat, 28.1% carbohydrate) in PaCO₂ levels during weaning from the ventilator, respect to a standard enteral nutrition (30% fat, 53.3% carbohydrate) (58). Another recent study on 51 critically ill ventilated children with pulmonary disease observed that the low-carbohydrate high-fat diet (30% carbohydrate, 50% fat) was effective in reducing carbon dioxide tension but did not reduce the duration or level of ventilatory support (59).

Similarly to healthy subjects and respiratory patients in outpatients conditions, the acute carbohydrate loading has detrimental effects on patients with acute respiratory distress (60, 61) and acute respiratory failure (43, 62, 63), acting as a precipitating factor.

Acute respiratory failure was reported in 3 patients under ventilatory support within hours after the beginning of total parenteral nutrition, probably due its high carbohydrate content. The substantial increase of the carbon dioxide production in these patients unable to increase their ventilatory response, led to development of hypercapnia and respiratory acidosis (43).

Ketogenic diet should be considered in severely obese people as a preventive measure for SARS-CoV-2 infection, in order to achieve a fast weight loss preserving lean mass. In addition, KD use may be hypothesized as an adjuvant therapy in obese infected patients.

Data obtained in respiratory patients, mainly lean, are indicative of the safety of low-carbohydrate high-fat diets in respiratory compromised patients, and some beneficial effects on respiratory parameters were recorded. KD administration may be helpful for obese patients with chronic hypercapnia, thanks to the reduced CO₂ production induced by the diet. Many respiratory patients are malnourished, and obese patients themselves are frequently sarcopenic, therefore an adequate protein supplementation is mandatory, since malnutrition may worsen the general prognosis. Many supplements for malnourished patients are rich in carbohydrates, and this is detrimental for their respiratory function, as extensively seen above. KD may be useful for obtaining an adequate protein intake, reducing the ventilatory requirements, the dyspnea and the risk of muscle fatigue. Anyhow, the studies are pretty old, diet administration quite short, samples relatively small, and ketosis not always addressed. In addition, we focused on chronic respiratory diseases, and not on acute infective respiratory illnesses, which may present other issues. Also active infections, in fact, are considered among the contraindications for VLCKD.

Anyway, KD has anti-inflammatory effects and may reduce the risk of cytokine storm, thanks to the anti-inflammatory effects of ketone bodies, and may have a direct anti-viral effect. Therefore, ketogenic diet may be an effective adjuvant therapy in obese non-critically ill COVID-19 patients, and may even be considered in respiratory lean patients. New focused clinical trials with adequate sample sizes, led by a multidisciplinary experienced team of pneumologists, endocrinologists and nutritionists, are needed to confirm the safety and the beneficial effects on ventilatory parameters of such approach in respiratory patients.

EG, LG, and CL: conceptualization. EG and RT: writing—original draft preparation. SM, AL, LG, and CL: writing—review and editing. All authors have read and agreed to the published version of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.