COVID-19 and abnormal uterine bleeding: potential associations and mechanisms.

Early in the COVID-19 pandemic it became apparent that there were differences in the response to SARS-CoV-2 infection in males and females. Examination of COVID-19 reported deaths from 16 countries in a 6-week period in 2020 revealed mortality rates from COVID-19 were 77% higher in men than in women (IRR = 1.77, 95%CI = 1.74, 1.79) [ 119 ]. This is likely to be due to a combination of gender-specific behaviours/activities/roles and biological sex differences. A Swedish cohort of post-menopausal women who tested positive for SARS-CoV-2 was divided into three categories for analysis: [ 1 ] decreased systemic estrogen levels [ 2 ], increased estrogen levels and [ 3 ] controls [ 120 ]. This revealed an absolute risk of death of 4.6% for the control group vs 10.1% and 2.1%, for the decreased and increased estrogen groups, respectively. This was despite adjustment for age, annual disposable income, highest level of education and the weighted Charlson Comorbidity Index. Further evidence for the reproductive hormones altering COVID-19 response comes from analysis of electronic health records in a large, international COVID-19 cohort, grouped by age. Pre-pubertal males and females had the same risk of infection and fatality rate, adult pre-menopausal women had a significantly higher rate of infection than men of the same age but lower fatality rate and post-menopausal women had a similar severe infection rate as men in the same age category [ 121 ]. Subanalysis of women over 50 using estradiol containing hormone replacement therapy revealed a 50% reduction in fatality rate [ 121 ]. In addition, a population based matched cohort study of female users of the COVID Symptom Study application in the UK revealed menopausal women had significantly higher rates of predicted COVID-19 and combined oral contraceptive pill (COCP) users had lower rates of predicted COVID-19 and reduced hospital attendance ( P =0.023). Taken together, these studies are consistent with a protective effect of the female reproductive hormones on serious SARS-CoV-2 infection and mortality ( Figure 3 ). The potential role of ovarian hormones, estrogen (E 2 ) and progesterone (P 4 ) on the development and symptomatology of acute COVID-19 and long COVID (IL, interleukin; CCL2, chemokine ligand 2; TNF, tumour necrosis factor; Treg: regulatory T cells). One mechanism underpinning these sex differences in acute COVID-19 outcomes may be the decreased innate immune inflammatory response and enhanced immune tolerance and antibody production in the presence of higher levels of estradiol and progesterone [ 6 ] ( Figure 3 ). It has long been known that females develop a heightened immune response to viruses than males [ 122 ]. Estradiol and progesterone have known anti-inflammatory and immunomodulatory actions [ 6 , 123–126 ], including suppression of the production of proinflammatory cytokines and chemokines (e.g. interleukin 6, interleukin 1β, TNF-α and CCL2), stimulation of CD4+ T-helper cell production of anti-inflammatory cytokines, and increase Th2-type anti-inflammatory responses ( Figure 3 ). Estradiol and progesterone can also enhance the expansion of regulatory T cells to promote immune tolerance. Collectively, these alterations may help to prevent or limit the cytokine storm associated with severe COVID-19. In addition to its impact on acute COVID-19, there is mounting evidence that the menstrual cycle may affect the development and symptomatology of long COVID ( Figure 3 ). A UK study of 327 hospitalised patients found that women under 50 were five times less likely to report feeling recovered at least 90 days following symptom onset when compared with men of the same age [ 51 ], suggesting the ovarian sex hormones may play a role in the pathogenesis of long COVID [ 127 ]. An online survey of 3762 people from 56 countries experiencing symptoms of COVID-19 lasting over 28 days included 2348 women <50 years old. More than a third of menstruating participants experienced relapses of their long COVID symptoms during or before menstruation [ 50 ]. Many patients with long COVID describe symptoms consistent with postural orthostatic tachycardia syndrome (POTS). In a study carried out prior to the COVID-19 pandemic, women with POTS were shown to have lower cardiac output and stroke volume and greater total peripheral resistance in the early proliferative phase of the cycle than mid secretory phase [ 128 ]. No such variation in parameters was observed in normal controls across the menstrual cycle. Hence, ovarian sex hormones may provide a therapeutic option for long COVID symptoms in some women and their safety profile and effectiveness should be investigated further ( Figure 3 ). Treatments targeting these ovarian hormones have been investigated as a treatment for acute COVID-19. A phase 2 randomised, double-blinded, placebo-controlled, multicentre trial evaluated the efficacy and safety of raloxifene treatment (a second-generation estrogen receptor modulator) for patients with mild-to-moderate COVID-19 [130]. The proportion of participants with undetectable SARS-CoV-2 after 7 days of treatment with raloxifene 60 mg and 120 mg was lower compared to placebo, RD = 0.37 (95% CI: 0.09 to 0.59) and RD = 0.22 (95% CI: −0.03 to −0-.45), respectively. There was no effect on requirement for supplemental oxygen or mechanical ventilation. Another phase 2, randomised, double-blind, placebo-controlled trial investigated the efficacy, safety and tolerability of estetrol (E4) versus placebo in hospitalised patients with SARS-CoV-2 [131]. The primary endpoint of improvement in recovery at day 28 was not altered between groups.

In recent years, there has been substantial media attention and public concern regarding menstrual disturbance during the COVID-19 pandemic. This may be due to COVID-19 vaccination, infection with the SARS-CoV-2 virus, pandemic-related stress and/or lifestyle changes, yet the independent contribution of each factor to the menstrual cycle is just beginning to be delineated [ 1–3 ]. This limits our ability to provide evidence-based advice to women and those who menstruate about the impact of COVID-19 on menstruation [ 4 ]. For many women, menstruation is an outward sign of fertility and health, and menstrual disturbances due to COVID-19 may provoke anxiety and stress as well as concern about the perceived risks of immunisation [ 5 ]. Further, the menstrual cycle and ovarian sex hormones have the potential to modulate disease susceptibility and severity [ 6 , 7 ], but our understanding of how the menstrual cycle impacts on SARS-CoV-2 infection and longer-term symptoms of COVID-19 also remains limited. This limited understanding of the dual relationship between COVID-19 and menstruation contrasts with the increasing awareness amongst healthcare professionals that the menstrual cycle can be used as a vital sign of female health [ 8 , 9 ] and the greater appreciation among scientists that sex should be considered in immunological studies [ 10 ], emphasising the importance of identifying associations and delineating mechanisms in this area. In this review, we focus on the impact of COVID-19 on menstrual bleeding, rather than its effects on fertility, dysmenorrhoea or other menstrual symptoms which have been the subject of previous reviews [ 7 , 11 , 12 ]. First, we define typical and abnormal menstrual parameters. Second, we examine the current evidence for and against an association between either SARS-CoV-2 infection, COVID-19 vaccination, longer-term symptoms of COVID-19 or psychological aspects of the COVID pandemic and menstrual disturbance. Third, potential mechanisms causing such associations are discussed, examining our knowledge of how COVID-19 impacts the hypothalamic–pituitary–ovarian (HPO) axis and the consequent effects on endometrial function as well as examination of potential local endometrial dysfunction. We draw on what is known from the study of SARS-CoV-2 pathophysiology in other systems. For example, SARS-CoV-2 uses the angiotensin-converting enzyme 2 (ACE2) and the cell protease type II transmembrane serine protease (TMPRSS2) to bind to the cell and for virus–cell fusion in lung tissue [ 13 , 14 ] and we discuss the potential role in reproductive tissue. Lastly, we review associations between the menstrual cycle and symptoms and outcomes of COVID-19, discussing potential mechanisms. Throughout, sources of bias within existing studies are evaluated before signposting recommendations for future research in this area. We acknowledge that not all women menstruate and that some people who do not identify as women do. Therefore, we use a combination of ‘women’ and ‘people who menstruate’ throughout this review.

Iatrogenic causes of AUB include use of anti-coagulant medications, gonadal steroids (e.g. estrogens, progestins, androgens) and nonsteroidal pharmaceuticals that contribute to ovulatory disorders (e.g. phenothiazines and tricyclic antidepressants) [ 15 ]. It is important to remain cognisant that treatments for COVID-19 may have off target effects that result in menstrual disturbance. For example, some of the common anti-viral medications used to treat early COVID-19 symptoms contain cytochrome P450 inhibitors (e.g. ritonavir). Concomitant use of hormonal contraception and drugs that inhibit cytochrome P450 has the potential to result in increased exposure to contraceptive hormones and increased side effects. This is a potential mechanism for reports of menstrual disturbance in those using hormonal contraception and who received treatment with anti-viral medications ( Figure 2 B), although these numbers are likely to be small. In patients hospitalised with COVID-19, low-dose corticosteroids in the form of dexamethasone have been shown to reduce mortality in those requiring oxygen or ventilatory support [ 110 ]. Endogenous glucocorticoids have been shown to inhibit endometrial angiogenesis [ 111 ] and regulation of endometrial blood vessel function is required to limit endometrial bleeding and menstrual blood loss [ 17 , 112 ]. A recent response-adaptive randomised placebo-controlled dose-finding parallel group trial revealed that low dose dexamethasone taken in the luteal phase of the menstrual cycle reduced menstrual blood loss [ 113 ]. Exposure to exogenous dexamethasone at other phases of the cycle may affect endometrial vasculature and represent a potential mechanism of reported menstrual disturbance in those with severe acute COVID-19 ( Figure 2 B). Given scientific research on long COVID is in its infancy, many of those suffering with longer term symptoms of COVID-19 are turning to off-label treatment options. There are many anecdotal reports of hyperbaric oxygen therapy (HBOT) improving long COVID symptoms. One study found statistically significant improvements in fatigue and cognitive scoring in 10 participants with long COVID after 10 sessions of HBOT to 2.4 atmospheres over 12 days [ 114 ]. This study was small and had no control group. A small ( n =73) randomised, placebo-controlled, double-blind trial has also shown benefits of daily HBOT [ 115 ] and a phase II clinical trial is in progress [ 116 ]. During HBOT, the arterial O 2 tension typically exceeds 1500 mmHg, and levels of 200–400 mmHg occur in tissues [ 117 ], consistent with it having an impact on endometrial tissue oxygen levels. Mice placed in hyperoxic conditions following progesterone withdrawal during simulated menstruation displayed less endometrial hypoxia and delayed endometrial repair [ 63 ]. The effects of intermittent HBOT on endometrial function remain undetermined, but prevention of physiological endometrial hypoxia has the potential to result in HMB [ 63 ]. Alternatively, it has been proposed that intermittent exposure to hyperoxic conditions and the relative changes in oxygen availability paradoxically stabilises HIF protein [ 117 , 118 ] and promotes angiogenesis. Given the presence and role of endometrial hypoxia in driving repair at menstruation, it is important to determine the impact of HBOT on menstrual symptoms and physiology. In addition, these studies highlight the importance of COVID-19 treatments as confounders when examining the impact of COVID-19 on menstruation. These hypothesised non-structural mechanisms that may underpin abnormal uterine bleeding with COVID-19 (AUB C/O/E/I, Figure 2 ) require investigation to ensure specific, evidence-based management. Although currently speculative, the FIGO AUB system 2 [ 15 ] provides a useful framework to design future research in this area, combining our knowledge of COVID-19 effects on other systems.

Taken together, the studies discussed above provide evidence of an association between acute COVID-19/long COVID and menstrual disturbance. However, methodological challenges (including self-reporting bias, small or non-representative sample sizes, cross-sectional data, and confounding factors) mean that it is currently unclear whether these associations represent causal relationships. There is some evidence that pandemic related stress did not result in significant population-level changes to ovulation or menstruation [ 53 ], consistent with a more specific impact of SARS-CoV-2 infection on menstruation. The acute menstrual disturbance described with COVID-19 illness/SARS-CoV-2 infection makes a non-structural cause for AUB more plausible than structural causes. In this section, we outline the physiology of menstruation before exploring the potential for abnormal uterine bleeding due to non-structural causes, specifically: coagulation disturbance (AUB-C), ovulatory dysfunction (AUB-O), disturbance of endometrial function (AUB-E) and iatrogenic causes (AUB-I) after SARS-CoV-2 infection.

An understanding of menstrual physiology is essential in delineating the mechanisms that may lead to AUB in SARS-CoV-2. Menstruation is the breakdown and shedding of the luminal portion of the endometrium, which occurs in the absence of pregnancy. Endometrial function is tightly regulated by the hypothalamic–pituitary–ovarian axis. Pulsatile secretion of gonadotrophin-releasing hormone (GnRH) from the hypothalamus acts on the pituitary to release follicle stimulating hormone (FSH) and luteinising hormone (LH), which stimulate the development of ovarian follicles and the secretion of oestradiol and progesterone. Oestradiol is the dominant hormone in the proliferative or follicular phase. Progesterone is released from the corpus luteum after ovulation and is the principal hormone during the secretory or luteal phase. Their actions on the endometrium result in complex interactions between the endocrine, vascular and immune systems. Progesterone and oestradiol levels decline sharply if implantation does not occur and the corpus luteum regresses during the late secretory stage. This results in local endometrial tissue inflammation with an influx of a host of inflammatory chemokines and cytokines and subsequent leukocytes [ 54–56 ]. This endometrial inflammation culminates in breakdown of the endometrium by matrix metalloproteinases (MMPs) and menstrual bleeding [ 57–59 ]. Endometrial tissue repair is then required to limit blood loss and maintain endometrial function in subsequent cycles. This endometrial repair occurs at the time of menstruation, with adjacent areas of endometrium undergoing breakdown and repair [ 60 ]. A transient, localised tissue hypoxia has been detected in the endometrium at menstruation [ 61 , 62 ] and was required for timely repair of the denuded endometrial surface [ 63 ]. A functional coagulation system is also required to limit menstrual blood loss, with the clotting cascade and platelet function important for achieving haemostasis post menstruation ( Figure 2 ) [ 64 , 65 ].

A lack of robust menstrual data during the COVID-19 pandemic made it difficult for the public and health professionals to determine the impact of acute SARS-CoV-2 infection, long COVID or COVID-19 vaccination on menstruation. We have reviewed available evidence and highlight that many studies from this time are limited by lack of standardised data collection, recall bias, variable sample sizes and confounders. Despite these limitations, collectively these studies suggest that SARS-CoV-2 infection (acute COVID-19 and long COVID) is associated with reported changes in menstrual bleeding parameters. Accurate information on the relationship between COVID-19 vaccination and menstrual disturbance is essential to avoid misinformation leading to vaccine hesitancy. Studies reviewed here report that some women may experience a small increase in cycle length or a change in bleeding pattern in the cycles following vaccination, but that this impact is transient and not specific to one brand of vaccine. In addition, this review discusses potential mechanisms involved in COVID-19 related menstrual disturbance. Combining our understanding of AUB pathways with research on the pathogenesis of COVID-19 in non-reproductive systems highlights a range of potential mechanisms that require further investigation. COVID-19 may disrupt menstrual regulation due to an impact on the HPO axis or direct effects on immune function at level of the endometrium. We acknowledge that there is limited data available to confirm these proposed mechanisms and this presents an important area for further targeted study. The number of people reporting symptoms of long COVID is increasing worldwide and there is growing interest to determine its impact across all body systems. Evidence presented here highlights that the menstrual cycle can impact long COVID symptomatology. This is an area which deserves further exploration to increase our understanding and develop better management strategies to improve quality of life. The reporting of menstrual disturbance from COVID-19 generated public interest and concern that resulted in vaccine hesitancy. These concerns were exacerbated by a lack of clear explanations from the scientific and medical community due to a lack of data collection. Moving forward, data and research on COVID-19 and menstruation will inform policy and will be invaluable for future vaccine development and understanding post-viral syndromes. Menstrual data must be considered as an important health parameter and should be included in future studies of long COVID, other viruses and vaccination trials. This will ensure that women and people who menstruate can remain informed about the expected impacts of viral infection on menstrual bleeding and that we develop our understanding of the relationship between menstruation and health outcomes.

Normal or typical menstrual parameters have been defined by the International Federation of Gynecology and Obstetrics (FIGO) Abnormal Uterine Bleeding (AUB) System 1. Menstrual bleeding typically has a frequency of every 24–38 days, duration of no more than 8 days, variation in shortest to longest menstrual cycle of less than or equal to 7–9 days (age dependant) and a flow volume of subjectively normal [ 15 ] ( Figure 1 ). The symptoms of abnormal uterine bleeding (AUB), affecting 1 in 4 women in society [ 16 , 17 ], are debilitating and can result from frequent/infrequent, prolonged, irregular or heavy menstrual bleeding [ 15 ]. Summary of the symptoms of abnormal uterine bleeding (AUB), as defined by the FIGO AUB System 1 [ 15 ], and the different aspects of COVID-19 which may be associated with reported menstrual changes. The symptom of heavy menstrual bleeding (HMB) is defined as excessive menstrual blood loss that interferes with a woman’s physical, emotional, social and/or material quality of life [ 18 ]. Depending on the population studied and definition adopted, global figures for HMB vary but are consistently high [ 19 , 20 ]. One in three women have been reported to find their menstrual loss excessive, with this figure rising to one in two as the menopause approaches [ 21 ]. Prior to the COVID-19 pandemic, over 800,000 women sought treatment for HMB per year in the UK alone [ 18 ] with many more suffering in silence [ 19 ]. Assessment of the economic impact of menstrual complaints in the United States (US) prior to the COVID pandemic demonstrated financial losses of greater than $2000 per patient per year due to absence from work and home management costs [ 22 ]. Therefore, any additional impact of COVID-19 on menstrual symptoms must be quantified to allow sufficient allocation of resources to address clinical needs. The underlying cause(s) of AUB may be classified as structural or non-structural, as outlined in the International Federation of Gynecology and Obstetrics (FIGO) AUB System 2 [ 15 ]. Structural causes are those that can usually be detected during routine examination or investigations (e.g., imaging or histopathology) and include Polyps, Adenomyosis, Leiomyomas (fibroids) or Malignancy (PALM) ( Figure 2 A). Non-structural causes are not detected on imaging but diagnosed by a detailed clinical history and examination, sometimes supported by laboratory tests. These non-structural causes include coagulopathies, ovulatory disorders, primary endometrial disorders, iatrogenic causes and those that are not otherwise classified (COEIN) ( Figure 2 A). Identifying the mechanisms underlying menstrual disturbance during COVID is crucial for effective, evidence-based management to maintain menstrual health. ( A ) Structural causes of AUB including Polyp, Adenomyosis, Leiomyoma and Malignancy (PALM). Non-structural causes of AUB including Coagulopathy, Ovulatory, Endometrial disorders, Iatrogenic or Not otherwise classified (COEIN) as defined by the FIGO AUB System 2 [ 15 ]. ( B ) The hypothesised mechanisms of menstrual disturbance based on reported systemic COVID-19 effects. Focusing on COEI, these effects may impact the endometrial processes required for normal menstruation and repair.