Impact of evening carbohydrate intake on the resolution of persistent night sweats in patients with long COVID: a case series

Introduction

Carbohydrates are the most abundant macromolecules on our planet and play a central role in cell metabolism, providing a rapid source of energy. Following carbohydrate intake, our body stores excess glucose in the form of glycogen, either in skeletal muscles (~500 g) for local consumption during exercise, in the liver (~100 g) to maintain blood glucose homeostasis during periods of daily fasting and stress,¹ or to a minor extent in the brain (~1 g) to provide lactate during the memory process.²

Imbalanced blood glucose levels lead to metabolic disorders such as diabetes, cardiovascular diseases, inflammatory diseases, or cancer. In addition, infectious diseases like COVID may deregulate energy homeostasis of patients by altering their mitochondria and, consequently, the amount of energy their cells can retrieve.^3–5

In such cases, when mitochondria are damaged, cells rely on glycolysis instead of aerobic oxidation. However, due to the lower yield of glycolysis (2 ATP instead of 30-38 ATP) higher amounts of glucose are required to maintain normoglycemia, resulting in a much faster depletion of glycogen stores.

Among COVID patients, about 20% of them experience symptoms several months after acute infection that can last for an indefinite period. This form of disease is colloquially referred to as “long COVID” and usually affects women, regardless of their need to be hospitalized or not.⁶ The most common symptoms reported by these patients are fatigue, headache, dyspnea, and brain fog, which relapse with physical or mental activity and stress.^7–9 Among systemic manifestations, an international cohort study revealed that 25% of 3,762 participants with long COVID experienced night sweats without a history of the condition.⁷ Similarly, another cohort of 5,163 patients with persistent COVID symptoms reported a night sweat prevalence of 28%, with a mean duration of 69 days and almost half of them still ongoing.⁸ Although night sweats can disappear over time, they cause significant discomfort and have not been attended in COVID patients.

Since night sweats are common in glycogen storage diseases and in diabetic patients with poor glucose control, we reasoned that glycogen depletion due to mitochondrial damage could be the underlying cause of persistent night sweats in patients with long COVID, and that replenishment of glycogen stores would revert them.

Here we describe three case reports of non-diabetic patients with long COVID whose persistent night sweats disappeared after increasing the carbohydrate content in their diets before bedtime. To our knowledge, this is the first time in the literature that persistent night sweats are addressed and resolved in COVID patients.

Methods

Persistent night sweats associated with long COVID were defined as having night sweats that required changing bedding (waking up with soaked sheets) even when it was not hot in the bedroom and that did not occur before SARS-CoV-2 infection.

Long COVID patients over 18 years of age, who were experiencing persistent night sweats for more than one month, and had no diabetes or previous history of night sweats were invited to participate in the study.

Volunteers were asked to report their night sweats on a Likert scale, ranging from 1 (sheets “completely dry” upon awakening) to 5 (sheets “soaking wet” upon awakening), the week before the intervention and daily during the 7-day diet intervention, along with their symptoms and exact diet. Agreed dietary interventions consisted of including carbohydrate-rich foods of the choice of the patients to their dinners. Sociodemographic characteristics, clinical conditions, and medications were described by the patients or consulted through access to their clinical histories, when possible.

Results

Only three patients participated in the study. All three shared a low evening carbohydrate intake (no dinner, early dinner, low-carb diet) before the nutritional intervention. Their characteristics are described in Table 1 and below in the text. Five volunteers backed out and decided not to join the study due to fear of weight gain once they learned about the nature of the intervention. One patient had a worsening of her symptoms before starting the intervention and declined to participate.

Case 1: High-protein, low-carb, gluten-free diet

Patient 1 was a 42-year-old Caucasian female with dysgeusia and anosmia as her first symptoms of SARS-CoV-2 infection. Two and a half months after testing positive by PCR she started showing night sweats in the absence of fever that soaked the sheets. No alcohol or illegal drug use was declared. When asked about previous medical conditions, she reported anxiety, depression, recurrent miscarriages, mild polycystic ovarian syndrome (PCOS), and irritable bowel syndrome (IBS). Consequently, regular use of antidepressants and anxiolytics in the past without concomitant sweating, and occasional use of antipyretics, were disclosed. She had taken low-dose metformin (250 mg/day) intermittently between 2011 and 2014 trying to become pregnant, given the diagnosis of mild PCOS, but she had no history of diabetes and her fasting blood glucose level was 93 mg/dL as tested after COVID-19 diagnosis. A wide variety of persistent long COVID symptoms were mentioned (anosmia, dysgeusia, chest pain, sore throat, cough, fatigue, neuropathy, muscle pain, insomnia, chills, skin rash, anxiety, headache, confusion, tinnitus, diarrhea, pre-atrial contractions, and high blood pressure).

She started the dietary intervention after two and a half months of persistent night sweats. At admission to the study, the patient only took Metoprolol (25 mg/day) to treat her pre-atrial contractions and Alprazolam (0.5 mg/day) for the anxiety. She had a mean of 5 out of 5 (soaking wet sheets) on the night sweat scale the week before modifying her diet. After the introduction of gluten-free macaroni on her dinners (Table 2), patient 1 experienced a progressive decline in her symptomatology, reaching completely dry sheets (1 out of 5) on the sixth and seventh days (Figure 1).

Table 2. Diet during the intervention (dinner).

In italics, food that was introduced.

Variable	Patient 1	Patient 2	Patient 3
Day 1	Salmon, salad of mixed greens with cucumber, carrot and avocado + a large serving of gluten-free macaroni and cheese	Sandwich with cooked ham, and cereals with skimmed milk	A piece of salmon an vegetables + three slices of brown bread with a boiled egg (later)
Day 2	Tilapia and green beans + a large serving of gluten-free macaroni and cheese	Sandwich with cooked ham, and cereals with skimmed milk	Two large slices of pizza + three slices of bread with a boiled egg (later)
Day 3	Lentil soup + a large serving of gluten-free macaroni and cheese	Sandwich with cooked ham, and cereals with skimmed milk	A pork fillet with carrots, one potato and a turnip + three slices of bread with a boiled egg (later)
Day 4	Salad of mixed greens with grilled steak and avocado + a large serving of gluten-free macaroni and cheese	Sandwich with cooked ham	Home-made vegetable soup + one portion of fried potatoes and onions (later)
Day 5	Salad of mixed greens with chicken and avocado + a large serving of gluten-free macaroni and cheese	Sandwich with quail eggs and bacon	Large serving of cod, with potatoes, carrots and broccoli + one portion of fried potatoes and onions (later)
Day 6	Salad of mixed greens with chicken and avocado + a large serving of gluten-free macaroni and cheese	Sandwich with semicured cheese	Chicken and pumpkin + one portion of fried potatoes and onions (later)
Day 7	Salad of mixed greens with chicken and avocado + a large serving of gluten-free macaroni and cheese	A bun with beef hamburger and fries	Two slices of pizza + one portion of fried potatoes and onions (later)

Figure 1. Night sweats reported by the participants before and after the nutritional intervention.

Likert scale ranging from 5 (soaking wet) to 1 (completely dry). The vertical dashed line shows the beginning of the intervention. The days before the diet modification are shown in negative.

Discussion

Night sweats can be caused by multiple factors such as infections, stress, cancer, medications, substance addiction, menopause or diabetes.¹⁰ However, none of the participants had a history of persistent night sweats prior to developing COVID.

Although the patients were women between 42-54 years of age, and that COVID-19 can elicit premature ovarian insufficiency,^11,12 no significant clinical or analytical signs suggestive of menopause were detected in any of the participants. Even if night sweats had been due to perimenopausal symptoms, they vanished after increasing their carbohydrate intake at night. On the other hand, medicines taken by Patient 1 (i.e. Alprazolam) and Patient 2 (i.e. Amplodipine, Simvastatin, Enalapril, Montelukast, and Diazepam) at the time of the intervention could potentially trigger night sweats in a very low percentage of patients (mainly women) after long periods of time; but the fact that they disappeared completely after the dietary intervention rules out this possibility. Patient 3 quit alcohol since being diagnosed with COVID-19, which could had played a role in her night sweats at the beginning, but not after 19 months, when the study took place.

Additionally, hypoglycemia is a common trigger in diabetic patients with poorly controlled glucose levels. However, none of the participants were diabetic, and their fasting glucose levels, accessed through medical records, were normal. Nevertheless, we hypothesized that faster depletion of the glycogen reservoirs in COVID patients could be the underlying cause of their persistent night sweats.

SARS-CoV-2 has been reported to alter mitochondrial metabolism^3–5 by reducing oxidative phosphorylation (yield: 30-38 ATP/glucose). This is compensated by a higher rate of glycolysis (yield: 2 ATP/glucose) and a greater dependence on glucose over glutamine and long-chain fatty acids (which require mitochondria for energy production). In consequence, cells need more molecules of glucose to achieve the same number of ATP molecules, which may precipitate premature exhaustion of the glycogen pools. Accordingly, postmortem biopsied hepatocytes from COVID-19 patients show damaged mitochondria and decreased glycogen granules,¹³ whereas mRNAs coding for proteins involved in glycogen biosynthesis are significantly downregulated in blood samples.¹⁴ Supporting this notion, continuous glucose monitoring of non-diabetic patients with mild COVID-19 revealed greater glycemic fluctuations and higher postprandial peaks accompanied by lower glycemic values throughout the night compared to healthy controls,¹⁵ which evidences an alteration of the liver glycogen metabolism.

Under normal conditions, during the overnight fast, metabolism gradually progresses to a catabolic state, promoting the breakdown of liver glycogen to support normoglycemia. At night, about half of the glucose supply is due to glycogenolysis, whereas the other half comes from gluconeogenesis in the liver and kidneys.^16,17

However, in COVID patients, the liver glycogen pool is smaller and the energy recovered by the cells relying on glycolysis is much less. This can lead to two possible scenarios. First, that faster depletion of the hepatic glycogen reservoir requires higher gluconeogenic rates to compensate, but these are not enough on their own to maintain constant normoglycemia, and hypoglycemic episodes develop at night. Second, that despite normoglycemic blood values are supported, cells do not get the needed energy output and patients display symptoms suggestive of hypoglycemia, despite normal test results. This condition is known as type I pseudohypoglycemia,¹⁸ and is due to lower energy obtained by cells with compromised mitochondrial function.⁴

In either case, epinephrine is released as part of the fight-or-flight response. Epinephrine is a potent stimulator of hepatic glucose output leading to a 2.5-fold increase in glucose production.¹⁹ It also triggers narrowing of the blood vessels and increased heart rates, raising the blood pressure and the apparent concentration of blood sugar to promote its delivery. In addition, it also facilitates breathing by relaxing the airways. On top of that, epinephrine can also activate adrenal receptors in the eccrine and apocrine glands, resulting in sweating.²⁰

Therefore, by increasing the carbohydrate intake of the participants right before going to bed, we were able to augment their hepatic glycogen pool and avoid the release of epinephrine and consequent night sweats. Curiously, all three shared a low evening carbohydrate intake (no dinner, early dinner, or low-carb diet) before the nutritional intervention.

It has been described that, under resting conditions, only 19% of the ingested glucose load is stored as glycogen in the liver, at a rate of approximately 6 g/h.²¹ This would explain the need to maintain carbohydrate supplementation for several days before the night sweats disappeared completely. In this study, we allowed participants to choose the carbohydrates of their choice pointing them the preference for starch sources over simple sugars, to avoid hyperglycemic peaks.

Epinephrine is secreted not only in case of hypoglycemia, but also under anxiety, stress, or shortness of breath. It is noteworthy that other symptoms reported by the patients may be due to either high or low epinephrine values (i.e. palpitations, high blood pressure, headaches, anxiety, insomnia) or hypoglycemic episodes (i.e. fatigue, difficulty focusing, confusion, memory loss).

Also compatible with our hypothesis is the fact that night sweats are more prevalent in women (all our cases were females). This may reflect a reduced hepatic glycogen content in females compared to males, as well as a reduced hepatic glucose output.^22,23

Although greater attention has been paid to high glycemic peaks, the greater risk of nocturnal hypoglycemia in COVID-19 patients has gone largely unnoticed by professionals. Current management guidelines for long-COVID symptoms are still sparse and imprecise, offering mainly holistic support or symptomatic treatments, but ignoring the underlying pathologies. Contrary to the general trend of thought, the endorsement of gradual physical activity and limited carbohydrate intake to improve overall patient condition may not be appropriate for everyone at all stages and may worsen some symptoms or contribute to their perpetuation.

Despite the limited number of patients included, our findings could be extrapolated to night sweats triggered by other diseases where mitochondrial impairment and faster hepatic glycogen depletion occur. We hope that future trials with more patients and resources confirm our results and the mechanisms proposed here. Unfortunately, fear of weight gain was the main reason to decline participation in the study.

Conclusions

Our study shows that persistent night sweats in long COVID patients can be resolved with dietary carbohydrate supplementation before bedtime. This is compatible with the hypothesis of rapid depletion of the hepatic glycogen store and insufficient cellular energy production due to mitochondrial damage in these patients.

This work also highlights the necessity to pay more attention to symptoms of hypoglycemia despite normal laboratory results in non-diabetic patients.

Ethical approval

The study was conducted in accordance with the Declaration of Helsinki, and approved by the Provincial Ethics Committee of Granada (code NS_23052021 approved on 02/02/2022).

Consent

Written informed consent for publication of their clinical details and/or clinical images was obtained from all the patients involved in the study.

Data availability

Data shared by the participants is confidential and not publicly available. However, researchers and reviewers may be granted access, once any personal information that could identify them is removed, and upon written request to the corresponding author, after explaining the reasons and stating that they will be treated confidentially and will be used only for research purposes.