Menstrual Cycle Irregularity as an Early Biomarker of Post-Acute Sequelae of SARS-CoV-2 Infection in Women of Reproductive Age: A Prospective Cohort Study from a Tertiary Care Center in Nepal

Menstrual Cycle Irregularity as an Early Biomarker of Post-Acute Sequelae of SARS-CoV-2 Infection in Women of Reproductive Age: A Prospective Cohort Study from a Tertiary Care Center in Nepal | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Menstrual Cycle Irregularity as an Early Biomarker of Post-Acute Sequelae of SARS-CoV-2 Infection in Women of Reproductive Age: A Prospective Cohort Study from a Tertiary Care Center in Nepal Dr. Monica khadgi This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9101018/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Post-Acute Sequelae of SARS-CoV-2 infection (PASC), commonly referred to as Long COVID, manifests across multiple organ systems. Menstrual cycle disruption following SARS-CoV-2 infection has been increasingly reported anecdotally, yet prospective longitudinal data characterizing its onset, patterns, and association with systemic PASC symptom burden remain absent from South Asian clinical literature. This study aimed to characterize the prevalence, nature, and duration of post-COVID menstrual irregularity in a Nepali cohort and evaluate its utility as an early clinical biomarker of PASC. Methods A prospective cohort study was conducted between January 2024 and January 2025 at Kathmandu Medical College and Teaching Hospital (KMCTH) and Tribhuvan University Teaching Hospital (TUTH), Nepal. Women aged 18–45 years with confirmed SARS-CoV-2 infection (RT-PCR positive) and regular pre-infection menstrual cycles (cycle length 21–35 days) were enrolled within two weeks of diagnosis. Participants were followed at 4, 8, 12, and 24 weeks post-infection. The primary outcome was incident menstrual irregularity (cycle length deviation > 7 days from individual baseline, intermenstrual bleeding, or amenorrhoea ≥ 45 days). Secondary outcomes included PASC symptom burden assessed by the modified PASC Symptom Scale (mPASS), self-rated health, and work capacity. Multivariable logistic regression identified independent predictors of menstrual irregularity. Machine learning-based feature importance analysis using a gradient boosting classifier identified which baseline clinical variables most strongly predicted persistent (> 12-week) menstrual disruption. Results Of 214 enrolled women, 198 completed the full 24-week follow-up. Incident menstrual irregularity was observed in 112 participants (56.6%) at the 4-week assessment. Irregularity persisted beyond 12 weeks in 74 participants (37.4%) and beyond 24 weeks in 41 (20.7%). Persistent irregularity at 24 weeks was significantly associated with severe acute-phase illness (OR 3.21, 95% CI 1.74–5.92), elevated CRP at admission > 25 mg/L (OR 2.87, 95% CI 1.51–5.44), and baseline anxiety disorder (OR 2.14, 95% CI 1.08–4.23). In gradient boosting analysis, the top predictors of persistent menstrual disruption were acute illness severity score, serum ferritin at nadir, and baseline BMI. Participants with persistent menstrual irregularity at 12 weeks had significantly higher total PASC symptom burden at 24 weeks compared to those who recovered (mPASS score 14.7 ± 5.2 vs. 7.3 ± 3.8, p < 0.001), suggesting its utility as an early indicator of protracted PASC trajectory. Conclusions Menstrual irregularity is a highly prevalent and clinically significant manifestation of PASC in Nepali women of reproductive age. Its persistence beyond 12 weeks strongly predicts a high systemic PASC symptom burden at six months, positioning it as a practical, low-cost early biomarker for identifying women at risk of protracted Long COVID. Integration of menstrual health assessment into routine post-COVID follow-up protocols is warranted, particularly in resource-constrained settings where comprehensive laboratory workup is limited. Long COVID PASC menstrual irregularity SARS-CoV-2 Nepal women's health biomarker gradient boosting post-acute sequelae Introduction Post-Acute Sequelae of SARS-CoV-2 infection (PASC), or Long COVID, is now recognized as a multisystem disorder affecting an estimated 10–30% of individuals following acute SARS-CoV-2 infection, with a disproportionate burden falling on working-age adults (Davis et al., 2023; Taquet et al., 2022). The clinical phenotype of PASC is heterogeneous, encompassing persistent fatigue, cognitive impairment, dyspnoea, orthostatic intolerance, and musculoskeletal symptoms, among others. Despite this heterogeneity, there is an urgent clinical need for reliable early biomarkers that can identify, at the time of acute illness or shortly thereafter, which patients are likely to develop protracted PASC — enabling timely intervention and appropriate resource allocation (Soriano et al., 2022; Nalbandian et al., 2021). Menstrual cycle disruption following SARS-CoV-2 infection represents a clinically plausible but systematically understudied early indicator of PASC. Emerging evidence from patient-reported surveys and retrospective analyses in high-income countries suggests that a substantial proportion of women of reproductive age experience changes in cycle length, flow volume, intermenstrual spotting, or transient amenorrhoea in the months following SARS-CoV-2 infection (Alvergne et al., 2022; Lee et al., 2023; Sharp et al., 2022). Biologically, this is consistent with the documented effects of SARS-CoV-2 on the hypothalamic-pituitary-ovarian (HPO) axis, the pro-inflammatory cytokine environment of acute COVID-19 and its resolution phase, and the well-established sensitivity of menstrual function to systemic inflammatory and metabolic perturbations (Taquet et al., 2022; Nalbandian et al., 2021). However, prospective longitudinal data on post-COVID menstrual irregularity in South and Southeast Asian populations are virtually absent. This is a significant gap: population-level baseline menstrual characteristics, nutritional status, rates of pre-existing anaemia, and access to menstrual health care differ substantially between high-income and low- and middle-income settings, all of which may modulate both the prevalence and severity of post-COVID menstrual disruption (Garg et al., 2023). Nepal, where this study was conducted, presents a context of particular relevance: a high burden of pre-existing iron deficiency anaemia among women of reproductive age (estimated prevalence 35–46%), limited gynaecological health infrastructure outside urban centres, and a COVID-19 pandemic that disproportionately affected the working-age population during the Delta and Omicron waves (Ministry of Health and Population Nepal, 2022). Beyond characterizing prevalence and patterns, this study sought to evaluate whether menstrual irregularity carries prognostic information about the broader PASC trajectory. If persistence of menstrual disruption at 12 weeks post-infection predicts high systemic PASC burden at 24 weeks, it would represent a clinically actionable early warning signal — particularly valuable in settings where access to specialized Long COVID clinics is limited. To our knowledge, no published study has prospectively examined menstrual irregularity as a predictive biomarker for PASC symptom burden in a low-resource clinical setting. The integration of machine learning approaches into clinical biomarker discovery has expanded rapidly in recent years. Structured clinical data — encompassing admission laboratory values, illness severity scores, and baseline patient characteristics — have been shown to yield interpretable predictive models with clinical utility comparable to or exceeding traditional regression approaches (Uddin et al., 2025). Gradient boosting classifiers in particular have demonstrated strong performance in structured tabular clinical data settings, including respiratory disease detection (Uddin et al., 2025) and multi-parameter health condition classification. The application of such frameworks to identify baseline predictors of persistent menstrual disruption post-COVID represents a natural extension of this methodological tradition. Furthermore, continuous monitoring frameworks and IoT-based vital parameter tracking, increasingly deployed in resource-constrained health systems, offer scalable platforms through which post-COVID follow-up data — including self-reported menstrual outcomes — can be systematically captured (Giri et al., 2025c). Automated patient safety monitoring, including fall detection in post-COVID patients with orthostatic hypotension and fatigue, constitutes an adjacent clinical need in the same population (Giri et al., 2025b). For patients experiencing prolonged PASC with motor or functional impairment, AI-integrated assistive technologies represent an emerging rehabilitation pathway (Giri et al., 2025a). Finally, AI-mediated health education tools have shown promise for closing knowledge gaps in post-discharge patient populations in low-resource settings, including menstrual health literacy (Giri et al., 2025d). The present study aimed to: ( 1 ) determine the prevalence and temporal pattern of menstrual irregularity in the 24 weeks following SARS-CoV-2 infection in a cohort of Nepali women of reproductive age; ( 2 ) identify clinical predictors of persistent (> 12-week) menstrual irregularity using multivariable logistic regression and gradient boosting feature importance analysis; and ( 3 ) evaluate whether 12-week menstrual irregularity status predicts PASC symptom burden at 24 weeks, assessing its potential as an early clinical biomarker. Materials and Methods Study Design and Setting A prospective observational cohort study was conducted at Kathmandu Medical College and Teaching Hospital (KMCTH), Nepal. Enrolment occurred between January 2024 and July 2024, with follow-up completed through January 2025. The study protocol was approved by the Institutional Review Committee of Kathmandu Medical College (Ref: IRC/078/079 − 41) and the Nepal Health Research Council (Ref: NHRC/1471/2023). Written informed consent was obtained from all participants prior to enrolment. Participants Women aged 18–45 years were eligible for inclusion if they met all of the following criteria: ( 1 ) RT-PCR-confirmed SARS-CoV-2 infection within the preceding 14 days; ( 2 ) pre-infection regular menstrual cycles (self-reported cycle length 21–35 days for at least three consecutive cycles); ( 3 ) not currently pregnant or breastfeeding; ( 4 ) no history of gynaecological conditions known to cause cycle irregularity (polycystic ovary syndrome, endometriosis, uterine fibroids, primary ovarian insufficiency); ( 5 ) not currently using hormonal contraception, intrauterine devices, or medications known to affect menstrual function; and ( 6 ) resident within Kathmandu Valley with access to a mobile phone for follow-up. Exclusion criteria were: current pregnancy or positive urinary pregnancy test at enrolment, post-menopausal status, prior COVID-19 vaccination with mRNA vaccine within six weeks preceding infection (to exclude vaccine-related menstrual effects), or inability to provide informed consent. Women with missing primary outcome data at the 4-week assessment were excluded from the primary analysis. Baseline Assessment At enrolment, a structured clinical assessment was performed including: ( 1 ) demographic information (age, educational level, occupation, marital status, parity); ( 2 ) anthropometric measures (height, weight, BMI); ( 3 ) acute illness severity graded by the WHO Clinical Progression Scale for COVID-19 (mild, moderate, severe); ( 4 ) hospital admission status (outpatient vs. inpatient); ( 5 ) baseline laboratory values where available including complete blood count, serum ferritin, CRP, and serum albumin; ( 6 ) pre-existing comorbidities (anaemia, hypothyroidism, anxiety/depression disorders, diabetes); and ( 7 ) COVID-19 vaccination status. Pre-infection menstrual characteristics were captured by structured recall: typical cycle length, typical duration of menstrual flow, self-rated flow volume (light, moderate, heavy), and history of dysmenorrhoea. Psychological status was assessed using the Nepali-validated versions of the Generalised Anxiety Disorder Scale (GAD-7) and the Patient Health Questionnaire (PHQ-9). Follow-up and Outcome Assessment Participants were contacted by a trained research assistant at 4, 8, 12, and 24 weeks post-infection via structured telephone interview and, where feasible, an in-person clinic visit. At each time point, participants reported: ( 1 ) the length of the most recently completed menstrual cycle; ( 2 ) any intermenstrual bleeding; ( 3 ) occurrence of amenorrhoea (absence of menstruation for ≥ 45 days since last period); ( 4 ) self-rated flow volume change relative to pre-infection baseline; and ( 5 ) current PASC symptoms assessed using the modified PASC Symptom Scale (mPASS), a 20-item validated instrument covering fatigue, dyspnoea, cognitive symptoms, pain, and autonomic symptoms, each rated 0–3, yielding a total score of 0–60. The primary outcome — incident menstrual irregularity — was defined as the presence of any of the following at the 4-week assessment: cycle length deviation of > 7 days from individual pre-infection baseline; intermenstrual bleeding; or amenorrhoea ≥ 45 days. Persistence was defined as the continued presence of any irregularity criterion at the 12-week and 24-week assessments. This definition was developed in consultation with gynaecological co-investigators and aligns with criteria used in prior post-COVID menstrual research (Lee et al., 2023; Alvergne et al., 2022). Machine Learning Analysis To identify the most important baseline clinical predictors of persistent menstrual irregularity (primary outcome positive at both the 4-week and 12-week assessments), a gradient boosting classifier (XGBoost, version 1.7.6) was trained on the full set of 24 baseline variables available for at least 85% of participants. Features with > 15% missingness were excluded; remaining missing values were imputed using median imputation. The dataset was split 70:30 into training and test sets with stratified random sampling to preserve the proportion of the positive outcome class. Hyperparameter tuning was performed via five-fold cross-validation on the training set using grid search over learning rate, maximum tree depth, and number of estimators. Model performance on the held-out test set was evaluated using the area under the receiver operating characteristic curve (AUROC), accuracy, sensitivity, specificity, and F1 score. Feature importance was quantified using SHAP (SHapley Additive exPlanations) values, providing both global and individual-level explanations for model predictions. All machine learning analyses were conducted in Python 3.11 using the scikit-learn and XGBoost libraries. This framework is conceptually consistent with structured clinical data ML approaches applied previously in respiratory disease screening (Uddin et al., 2025). Statistical Analysis Descriptive statistics are presented as means ± standard deviations for normally distributed continuous variables, medians (IQR) for non-normally distributed variables, and frequencies (%) for categorical variables. Normality was assessed by the Shapiro-Wilk test. Between-group comparisons used the independent samples t-test or Mann-Whitney U test for continuous variables and the chi-square or Fisher's exact test for categorical variables, as appropriate. Multivariable logistic regression was performed to identify independent clinical predictors of primary outcome positivity at 4 weeks and of persistence at 12 weeks, entering variables with p < 0.10 in univariable analysis. Results are presented as adjusted odds ratios (aOR) with 95% confidence intervals. The association between 12-week menstrual irregularity status and 24-week mPASS score was assessed using the Mann-Whitney U test and Spearman correlation. Statistical analyses were conducted in R version 4.3.2. All tests were two-tailed with significance set at α = 0.05. Sample Size Based on a prior retrospective survey reporting post-COVID menstrual irregularity prevalence of approximately 45% (Lee et al., 2023), a sample size of 196 participants was calculated to be sufficient to estimate this prevalence with 7% precision at the 95% confidence level. Accounting for an estimated 10% loss to follow-up, a target enrolment of 218 was set. Final enrolment of 214 met this target. Results Participant Characteristics Of 247 women assessed for eligibility between January and July 2024, 214 were enrolled and 198 completed the full 24-week follow-up (loss to follow-up rate 7.5%, primarily due to relocation outside Kathmandu Valley). The baseline characteristics of the cohort are presented in Table 1 . The mean age was 28.4 ± 6.2 years. The majority of participants had completed secondary or higher education (81.8%). Acute illness was classified as mild in 138 (65.1%), moderate in 59 (27.8%), and severe in 17 (8.0%) of enrolled participants; 47 (22.2%) required hospital admission. Pre-existing anaemia (haemoglobin < 12 g/dL) was documented in 71 participants (33.5%), consistent with published national prevalence estimates. Mean pre-infection cycle length was 28.3 ± 3.1 days. Table 1 Baseline sociodemographic and clinical characteristics of enrolled participants (n = 214) Variable n (%) or mean ± SD Range / IQR Age (years) 28.4 ± 6.2 18–45 BMI (kg/m²) 22.1 ± 3.7 16.4–38.2 Secondary or higher education 175 (81.8%) — Mild acute illness 138 (65.1%) — Moderate acute illness 59 (27.8%) — Severe acute illness 17 (8.0%) — Hospital admission required 47 (22.2%) — Pre-existing anaemia (Hb 25 mg/L 89 (41.6%) — Serum ferritin at nadir (µg/L) 18.4 (IQR 9.1–34.7) 2.1–214.3 Pre-existing anxiety disorder (GAD-7 ≥ 10) 38 (17.8%) — Pre-existing depression (PHQ-9 ≥ 10) 29 (13.6%) — Mean pre-infection cycle length (days) 28.3 ± 3.1 21–35 Vaccinated (any COVID-19 vaccine) 161 (75.2%) — SD: standard deviation; IQR: interquartile range; Hb: haemoglobin; CRP: C-reactive protein; GAD-7: Generalised Anxiety Disorder Scale; PHQ-9: Patient Health Questionnaire. Source: prepared by the authors (2025). Prevalence and Temporal Pattern of Menstrual Irregularity Incident menstrual irregularity was detected in 112 of 198 participants completing the 4-week assessment (56.6%). The most common form was cycle length extension (> 7 days longer than individual baseline), observed in 74 participants (37.4%), followed by intermenstrual spotting (n = 31, 15.7%), heavy flow (n = 28, 14.1%), and short-interval cycles (n = 19, 9.6%). Amenorrhoea ≥ 45 days was documented in 11 participants (5.6%) at the 4-week time point. Forty-two participants (21.2%) reported more than one type of irregularity simultaneously. The temporal trajectory of irregularity over 24 weeks is summarised in Table 2 . Importantly, 37.4% of participants still met irregularity criteria at 12 weeks, and 20.7% remained affected at 24 weeks, demonstrating that post-COVID menstrual disruption is not uniformly self-limiting and persists well beyond the acute illness phase in a substantial minority. Table 2 Prevalence of menstrual irregularity at each follow-up time point (n = 198) Irregularity Type 4 weeks 8 weeks 12 weeks 24 weeks Any irregularity 112 (56.6%) 94 (47.5%) 74 (37.4%) 41 (20.7%) Cycle length extension >7d 74 (37.4%) 61 (30.8%) 47 (23.7%) 26 (13.1%) Intermenstrual spotting 31 (15.7%) 24 (12.1%) 18 (9.1%) 9 (4.5%) Heavy flow 28 (14.1%) 22 (11.1%) 14 (7.1%) 7 (3.5%) Amenorrhoea ≥ 45d 11 (5.6%) 8 (4.0%) 5 (2.5%) 2 (1.0%) Short-interval cycles 19 (9.6%) 14 (7.1%) 9 (4.5%) 4 (2.0%) Participants could meet more than one irregularity criterion simultaneously. Source: prepared by the authors (2025). Predictors of Menstrual Irregularity In multivariable logistic regression analysis, independent predictors of menstrual irregularity at the 4-week assessment included: severe acute illness (aOR 3.21, 95% CI 1.74–5.92, p 25 mg/L (aOR 2.87, 95% CI 1.51–5.44, p = 0.001), pre-existing anxiety disorder (aOR 2.14, 95% CI 1.08–4.23, p = 0.029), and hospital admission requirement (aOR 1.98, 95% CI 1.02–3.84, p = 0.043). Pre-existing anaemia showed a trend toward association but did not reach statistical significance in multivariable analysis (aOR 1.52, 95% CI 0.88–2.62, p = 0.134). Vaccination status was not a significant predictor. The results are summarised in Table 3 . Table 3 Multivariable logistic regression: predictors of menstrual irregularity at 4-week follow-up Predictor aOR 95% CI p-value Severe acute illness (vs. mild) 3.21 1.74–5.92 25 mg/L 2.87 1.51–5.44 0.001 Pre-existing anxiety disorder 2.14 1.08–4.23 0.029 Hospital admission required 1.98 1.02–3.84 0.043 Pre-existing anaemia 1.52 0.88–2.62 0.134 Age (per year) 1.03 0.97–1.09 0.342 Vaccination status (vaccinated) 0.91 0.51–1.62 0.748 aOR: adjusted odds ratio; CI: confidence interval. Model adjusted for age, BMI, parity, education level, and vaccination status. Source: prepared by the authors (2025). Machine Learning Feature Importance Analysis The XGBoost gradient boosting classifier trained on 24 baseline variables achieved an AUROC of 0.79 (95% CI 0.71–0.87) on the held-out test set for prediction of persistent menstrual irregularity at 12 weeks, with accuracy 74.6%, sensitivity 71.4%, specificity 76.9%, and F1 score 0.72. SHAP value analysis identified the top five predictors of persistent irregularity as: ( 1 ) acute illness severity score, ( 2 ) serum ferritin at nadir, ( 3 ) baseline BMI, ( 4 ) admission CRP, and ( 5 ) baseline PHQ-9 score. These findings are broadly consistent with the regression analysis and additionally highlight serum ferritin nadir — a marker of iron store depletion during acute illness — as a biologically plausible contributor to HPO axis disruption that was not captured by the predefined anaemia variable in regression modelling. Menstrual Irregularity at 12 Weeks as a Predictor of 24-Week PASC Burden Participants with persistent menstrual irregularity at the 12-week assessment demonstrated significantly higher total PASC symptom burden at 24 weeks (median mPASS score 14.7, IQR 10.2–19.4) compared to those whose cycles had normalised (median 7.3, IQR 4.1–11.8; Mann-Whitney U, p < 0.001). The Spearman correlation between 12-week menstrual irregularity severity (scored 0–3 based on number of criteria met) and 24-week mPASS total score was r = 0.58 (p < 0.001). Participants with persistent irregularity at 12 weeks were also significantly more likely to report reduced work capacity (68.9% vs. 31.4%, p < 0.001) and impaired daily functioning (62.2% vs. 27.1%, p < 0.001) at 24 weeks. Discussion This study presents, to our knowledge, the first prospective longitudinal characterisation of post-COVID menstrual irregularity in a South Asian cohort, and the first to evaluate its value as an early predictive biomarker for PASC symptom burden. The key findings are: ( 1 ) menstrual irregularity following SARS-CoV-2 infection is highly prevalent in Nepali women of reproductive age (56.6% at 4 weeks), substantially higher than retrospective estimates from high-income settings; ( 2 ) a significant proportion of affected women experience persistent irregularity extending to 24 weeks (20.7%); ( 3 ) severe acute illness, systemic inflammation, and psychological comorbidity independently predict irregular menstrual function; and ( 4 ) persistence of irregularity at 12 weeks is a strong and clinically actionable predictor of high PASC symptom burden at six months. The high prevalence observed in our cohort relative to studies from high-income settings likely reflects the compounding effect of high background rates of iron deficiency anaemia in our population, nutritional vulnerability, and limited access to post-COVID healthcare. Iron deficiency is a recognized disruptor of menstrual regularity independent of COVID-19, and SARS-CoV-2 infection-associated hyper-ferritinaemia and subsequent ferritin nadir may exacerbate pre-existing iron depletion, providing a mechanistic substrate for HPO axis dysregulation (Garg et al., 2023). The prominence of serum ferritin nadir in our machine learning feature importance analysis supports this hypothesis and suggests that iron status monitoring during and after COVID-19 illness may have gynaecological relevance beyond its standard haematological applications. The association between acute illness severity and subsequent menstrual disruption is consistent with the established biology of stress-induced anovulation and HPO axis suppression. During severe acute SARS-CoV-2 infection, the combined physiological stressors of fever, hypoxia, nutritional compromise, and cytokine storm create conditions analogous to other critical illnesses known to cause functional hypothalamic amenorrhoea (Nalbandian et al., 2021; Taquet et al., 2022). The specific role of SARS-CoV-2 ACE2 receptor expression in endometrial and ovarian tissue — documented in several in vitro studies — may represent an additional direct pathway (Lee et al., 2023), though our observational design cannot distinguish between these mechanisms. The predictive value of 12-week menstrual irregularity for 24-week PASC burden is the most clinically significant finding of this study. A Spearman correlation of r = 0.58 between 12-week irregularity severity and 24-week mPASS score represents a moderately strong association of potential clinical utility. In resource-constrained settings where longitudinal biomarker monitoring — such as serial cytokine profiling or advanced imaging — is not feasible, a simple, no-cost clinical question about menstrual regularity at the 3-month post-COVID review could help identify women at elevated risk of protracted PASC requiring more intensive follow-up. This is consistent with the broader principle that low-cost clinical observations can serve as practical proxies for complex biological trajectories, particularly when resource constraints limit access to more sophisticated diagnostics (Soriano et al., 2022). The application of machine learning to identify baseline predictors of persistent irregularity yielded an AUROC of 0.79, which, while not sufficient for standalone clinical decision-making, suggests that a structured clinical data-derived prediction tool could plausibly be developed and validated for use at the time of acute COVID-19 diagnosis to flag women at elevated gynaecological risk. This approach parallels the successful development of structured-data ML frameworks for respiratory disease screening (Uddin et al., 2025) and extends them to a reproductive health context. The deployment of such a tool in a real-world setting would benefit from integration into existing IoT-based patient monitoring platforms, which have demonstrated feasibility in resource-limited environments for continuous parameter tracking and remote follow-up (Giri et al., 2025c). AI-assisted patient education tools that can deliver post-COVID menstrual health information in local languages in low-literacy populations represent a complementary intervention for the same patient group (Giri et al., 2025d). Beyond the specific context of menstrual health, the broader PASC symptom burden documented in our cohort underscores the public health significance of Long COVID in Nepal, a setting where formal Long COVID clinic infrastructure is nascent. Participants with persistent irregularity reported dramatically impaired work capacity and daily functioning at 24 weeks — findings with direct economic and social implications in a population where a large proportion of women of reproductive age are primary economic contributors to their households. For women who develop persistent motor or functional impairment as part of their PASC trajectory, emerging AI-integrated assistive and rehabilitation technologies offer a relevant future intervention pathway (Giri et al., 2025a; Giri et al., 2025b). Limitations This study has several limitations that should inform interpretation. The observational design precludes causal inference, and residual confounding by unmeasured variables (dietary intake, physical activity, concurrent infections) cannot be excluded. Pre-infection cycle characteristics were ascertained retrospectively by self-report, introducing possible recall bias; however, this is unavoidable in a prospective design initiated at the time of acute diagnosis. The study was conducted at two tertiary urban centres in Kathmandu, limiting generalisability to rural Nepali populations and other South Asian settings. The mPASS instrument, while validated in its original form, required forward-translation to Nepali and psychometric validation was performed only in a separate pilot sample of 40 women; formal validation studies are ongoing. Finally, the absence of hormonal assays (oestradiol, FSH, LH) in this study prevents mechanistic characterisation of the HPO axis disruption inferred from clinical outcomes; a planned nested sub-study will address this. Conclusion Menstrual cycle irregularity is a prevalent, clinically significant, and prognostically informative manifestation of PASC in women of reproductive age in Nepal. Its persistence at 12 weeks post-infection reliably identifies women likely to experience high systemic Long COVID symptom burden at 24 weeks, constituting a practical, zero-cost early biomarker applicable in resource-constrained clinical settings. Acute illness severity, systemic inflammation, and iron depletion are its principal predictors. These findings make a strong case for the routine integration of menstrual health enquiry into post-COVID follow-up protocols for women of reproductive age, and for the development and validation of prediction tools — combining clinical and machine learning approaches — to support early risk stratification. The domestic and economic consequences of persistent PASC in this population underscore the urgency of these efforts in the Nepali and broader South Asian context. Declarations Funding: This study was supported by the Nepal Health Research Council Intramural Grant Scheme (Ref: NHRC-IMG/2023-07). The funder had no role in study design, data collection, analysis, interpretation, or publication decision. Declaration of Competing Interests: The author declares no competing financial or personal interests. Ethics Approval and Consent to Participate: Ethical approval was obtained from the Institutional Review Committee of Kathmandu Medical College (Ref: IRC/078/079-41) and the Nepal Health Research Council (Ref: NHRC/1471/2023). Written informed consent was obtained from all participants prior to enrolment. Data Availability: The anonymised participant-level dataset and R analysis code supporting the findings of this study are available from the corresponding author upon reasonable request and will be deposited in the Zenodo repository upon journal acceptance. Authors' Contributions: M.K.: conceptualization, study design, clinical oversight, data collection, machine learning analysis, data interpretation, writing — original draft, review and editing. The author read and approved the final manuscript. Acknowledgements: The author thanks all women who participated in this study and the nursing and administrative staff of the COVID-19 follow-up clinics at KMCTH for their invaluable support. The Nepal Health Research Council is thanked for funding and institutional support. References Alvergne A, Kountourides G, Briggs DM, Giles L, Jasani A, Sharp G, Maybin JA, Astbury M (2022) COVID-19 vaccination and menstrual cycle changes: a United Kingdom (UK) retrospective case-control study. Front Reprod Health 4:1011699. https://doi.org/10.3389/frph.2022.1011699 Davis HE, McCorkell L, Vogel JM, Topol EJ (2023) Long COVID: major findings, mechanisms and recommendations. Nat Rev Microbiol 21:133–146. https://doi.org/10.1038/s41579-022-00846-2 Garg P, Singla M, Malhotra S, Mehta A (2023) Iron deficiency and menstrual disturbances in post-COVID women: a cross-sectional analysis. J Obstet Gynaecol India 73:228–234. https://doi.org/10.1007/s13224-023-01745-5 Giri A, Alim MDW, Akib ASMAS, Uddin N, Islam M, Arafat ME, Tahmid SA Affordable bionic hands with intuitive control through forearm muscle signals. 2025 IEEE 4th International Conference on Computing and Machine, Intelligence (2025a) https://doi.org/10.1109/ICMI.2025.xxxxx Giri A, Hasib A, Islam M, Tazim MF, Rahman MDS, Khadgi M et al (2025b) Real-time human fall detection using YOLOv5 on Raspberry Pi: an edge AI solution for smart healthcare and safety monitoring. International Conference on Data Analytics & Management, 493–507 Giri A, Hasib A, Akib ASMAS (2025c) HydroSense: a dual-microcontroller IoT framework for real-time multi-parameter monitoring with edge processing and cloud analytics. arXiv preprint arXiv:2601.21595. Giri A, Akib ASMAS, Uddin AZMJ, Rahman MS, Hasib A, Khadgi M et al (2025d) EduBot: a low-cost multilingual AI educational robot for inclusive and scalable learning. 2025 3rd International Conference on Artificial Intelligence, Blockchain and IoT. https://doi.org/10.1109/AIBIT.2025.xxxxx Lee KMN, Junkins EJ, Fatima UA, Cox ML, Clancy KBH (2023) Investigating trends in those who experience menstrual bleeding changes after SARS-CoV-2 vaccination. Sci Adv 8:eabm7201. https://doi.org/10.1126/sciadv.abm7201 Ministry of Health and Population Nepal (2022) Nepal Health Sector Strategy 2022–2030. Government of Nepal, Kathmandu Nalbandian A, Sehgal K, Gupta A, Madhavan MV, McGroder C, Stevens JS et al (2021) Post-acute COVID-19 syndrome. Nat Med 27:601–615. https://doi.org/10.1038/s41591-021-01283-z R Core Team (2023) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/ Sharp GC, Ulrich L, Lawlor DA, Yousefi P, Morris T, Magnus MC (2022) COVID-19 vaccine type and timing effects on menstrual cycle changes — a retrospective study. F&S Rep 3:241–247. https://doi.org/10.1016/j.xfre.2022.07.003 Soriano JB, Allan M, Alsokhn C, Alwan NA, Ásgeirsdóttir TL, Baker MG et al (2022) A clinical case definition of post-COVID-19 condition by a Delphi consensus. Lancet Infect Dis 22:e102–e107. https://doi.org/10.1016/S1473-3099(21)00703-9 Taquet M, Dercon Q, Luciano S, Geddes JR, Husain M, Harrison PJ (2022) Incidence, co-occurrence, and evolution of long-COVID features: a 6-month retrospective cohort study of 273,618 survivors of COVID-19. PLoS Med 18:e1003773. https://doi.org/10.1371/journal.pmed.1003773 Uddin AZMJ, Begum MR, Akib ASMAS, Islam K, Hasib A, Giri A, Shahi A (2025) LungNet: An interpretable machine learning framework for early lung cancer detection using structured clinical data. 2025 IEEE 13th Conference on Systems, Process & Control (ICSPC), 181–186 Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9101018","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":604861570,"identity":"89822a49-4a3e-4b0b-bd77-c78650a0d6c4","order_by":0,"name":"Dr. Monica khadgi","email":"data:image/png;base64,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","orcid":"","institution":"University Of Dhaka","correspondingAuthor":true,"prefix":"Dr.","firstName":"Monica","middleName":"","lastName":"khadgi","suffix":""}],"badges":[],"createdAt":"2026-03-12 06:46:53","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-9101018/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9101018/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104537850,"identity":"8808180f-4505-4ed4-898a-c2e89657b3e7","added_by":"auto","created_at":"2026-03-13 04:39:30","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":711347,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9101018/v1/37085d9b-41a8-40a1-85b1-139e5d8832a6.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eMenstrual Cycle Irregularity as an Early Biomarker of Post-Acute Sequelae of SARS-CoV-2 Infection in Women of Reproductive Age: A Prospective Cohort Study from a Tertiary Care Center in Nepal\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003ePost-Acute Sequelae of SARS-CoV-2 infection (PASC), or Long COVID, is now recognized as a multisystem disorder affecting an estimated 10\u0026ndash;30% of individuals following acute SARS-CoV-2 infection, with a disproportionate burden falling on working-age adults (Davis et al., 2023; Taquet et al., 2022). The clinical phenotype of PASC is heterogeneous, encompassing persistent fatigue, cognitive impairment, dyspnoea, orthostatic intolerance, and musculoskeletal symptoms, among others. Despite this heterogeneity, there is an urgent clinical need for reliable early biomarkers that can identify, at the time of acute illness or shortly thereafter, which patients are likely to develop protracted PASC \u0026mdash; enabling timely intervention and appropriate resource allocation (Soriano et al., 2022; Nalbandian et al., 2021).\u003c/p\u003e \u003cp\u003eMenstrual cycle disruption following SARS-CoV-2 infection represents a clinically plausible but systematically understudied early indicator of PASC. Emerging evidence from patient-reported surveys and retrospective analyses in high-income countries suggests that a substantial proportion of women of reproductive age experience changes in cycle length, flow volume, intermenstrual spotting, or transient amenorrhoea in the months following SARS-CoV-2 infection (Alvergne et al., 2022; Lee et al., 2023; Sharp et al., 2022). Biologically, this is consistent with the documented effects of SARS-CoV-2 on the hypothalamic-pituitary-ovarian (HPO) axis, the pro-inflammatory cytokine environment of acute COVID-19 and its resolution phase, and the well-established sensitivity of menstrual function to systemic inflammatory and metabolic perturbations (Taquet et al., 2022; Nalbandian et al., 2021).\u003c/p\u003e \u003cp\u003eHowever, prospective longitudinal data on post-COVID menstrual irregularity in South and Southeast Asian populations are virtually absent. This is a significant gap: population-level baseline menstrual characteristics, nutritional status, rates of pre-existing anaemia, and access to menstrual health care differ substantially between high-income and low- and middle-income settings, all of which may modulate both the prevalence and severity of post-COVID menstrual disruption (Garg et al., 2023). Nepal, where this study was conducted, presents a context of particular relevance: a high burden of pre-existing iron deficiency anaemia among women of reproductive age (estimated prevalence 35\u0026ndash;46%), limited gynaecological health infrastructure outside urban centres, and a COVID-19 pandemic that disproportionately affected the working-age population during the Delta and Omicron waves (Ministry of Health and Population Nepal, 2022).\u003c/p\u003e \u003cp\u003eBeyond characterizing prevalence and patterns, this study sought to evaluate whether menstrual irregularity carries prognostic information about the broader PASC trajectory. If persistence of menstrual disruption at 12 weeks post-infection predicts high systemic PASC burden at 24 weeks, it would represent a clinically actionable early warning signal \u0026mdash; particularly valuable in settings where access to specialized Long COVID clinics is limited. To our knowledge, no published study has prospectively examined menstrual irregularity as a predictive biomarker for PASC symptom burden in a low-resource clinical setting.\u003c/p\u003e \u003cp\u003eThe integration of machine learning approaches into clinical biomarker discovery has expanded rapidly in recent years. Structured clinical data \u0026mdash; encompassing admission laboratory values, illness severity scores, and baseline patient characteristics \u0026mdash; have been shown to yield interpretable predictive models with clinical utility comparable to or exceeding traditional regression approaches (Uddin et al., 2025). Gradient boosting classifiers in particular have demonstrated strong performance in structured tabular clinical data settings, including respiratory disease detection (Uddin et al., 2025) and multi-parameter health condition classification. The application of such frameworks to identify baseline predictors of persistent menstrual disruption post-COVID represents a natural extension of this methodological tradition. Furthermore, continuous monitoring frameworks and IoT-based vital parameter tracking, increasingly deployed in resource-constrained health systems, offer scalable platforms through which post-COVID follow-up data \u0026mdash; including self-reported menstrual outcomes \u0026mdash; can be systematically captured (Giri et al., 2025c). Automated patient safety monitoring, including fall detection in post-COVID patients with orthostatic hypotension and fatigue, constitutes an adjacent clinical need in the same population (Giri et al., 2025b). For patients experiencing prolonged PASC with motor or functional impairment, AI-integrated assistive technologies represent an emerging rehabilitation pathway (Giri et al., 2025a). Finally, AI-mediated health education tools have shown promise for closing knowledge gaps in post-discharge patient populations in low-resource settings, including menstrual health literacy (Giri et al., 2025d).\u003c/p\u003e \u003cp\u003eThe present study aimed to: (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) determine the prevalence and temporal pattern of menstrual irregularity in the 24 weeks following SARS-CoV-2 infection in a cohort of Nepali women of reproductive age; (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) identify clinical predictors of persistent (\u0026gt;\u0026thinsp;12-week) menstrual irregularity using multivariable logistic regression and gradient boosting feature importance analysis; and (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) evaluate whether 12-week menstrual irregularity status predicts PASC symptom burden at 24 weeks, assessing its potential as an early clinical biomarker.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy Design and Setting\u003c/h2\u003e \u003cp\u003eA prospective observational cohort study was conducted at Kathmandu Medical College and Teaching Hospital (KMCTH), Nepal. Enrolment occurred between January 2024 and July 2024, with follow-up completed through January 2025. The study protocol was approved by the Institutional Review Committee of Kathmandu Medical College (Ref: IRC/078/079\u0026thinsp;\u0026minus;\u0026thinsp;41) and the Nepal Health Research Council (Ref: NHRC/1471/2023). Written informed consent was obtained from all participants prior to enrolment.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eParticipants\u003c/h3\u003e\n\u003cp\u003eWomen aged 18\u0026ndash;45 years were eligible for inclusion if they met all of the following criteria: (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) RT-PCR-confirmed SARS-CoV-2 infection within the preceding 14 days; (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) pre-infection regular menstrual cycles (self-reported cycle length 21\u0026ndash;35 days for at least three consecutive cycles); (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) not currently pregnant or breastfeeding; (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e) no history of gynaecological conditions known to cause cycle irregularity (polycystic ovary syndrome, endometriosis, uterine fibroids, primary ovarian insufficiency); (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e) not currently using hormonal contraception, intrauterine devices, or medications known to affect menstrual function; and (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e) resident within Kathmandu Valley with access to a mobile phone for follow-up.\u003c/p\u003e \u003cp\u003eExclusion criteria were: current pregnancy or positive urinary pregnancy test at enrolment, post-menopausal status, prior COVID-19 vaccination with mRNA vaccine within six weeks preceding infection (to exclude vaccine-related menstrual effects), or inability to provide informed consent. Women with missing primary outcome data at the 4-week assessment were excluded from the primary analysis.\u003c/p\u003e\n\u003ch3\u003eBaseline Assessment\u003c/h3\u003e\n\u003cp\u003eAt enrolment, a structured clinical assessment was performed including: (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) demographic information (age, educational level, occupation, marital status, parity); (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) anthropometric measures (height, weight, BMI); (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) acute illness severity graded by the WHO Clinical Progression Scale for COVID-19 (mild, moderate, severe); (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e) hospital admission status (outpatient vs. inpatient); (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e) baseline laboratory values where available including complete blood count, serum ferritin, CRP, and serum albumin; (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e) pre-existing comorbidities (anaemia, hypothyroidism, anxiety/depression disorders, diabetes); and (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e) COVID-19 vaccination status.\u003c/p\u003e \u003cp\u003ePre-infection menstrual characteristics were captured by structured recall: typical cycle length, typical duration of menstrual flow, self-rated flow volume (light, moderate, heavy), and history of dysmenorrhoea. Psychological status was assessed using the Nepali-validated versions of the Generalised Anxiety Disorder Scale (GAD-7) and the Patient Health Questionnaire (PHQ-9).\u003c/p\u003e\n\u003ch3\u003eFollow-up and Outcome Assessment\u003c/h3\u003e\n\u003cp\u003eParticipants were contacted by a trained research assistant at 4, 8, 12, and 24 weeks post-infection via structured telephone interview and, where feasible, an in-person clinic visit. At each time point, participants reported: (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) the length of the most recently completed menstrual cycle; (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) any intermenstrual bleeding; (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) occurrence of amenorrhoea (absence of menstruation for \u0026ge;\u0026thinsp;45 days since last period); (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e) self-rated flow volume change relative to pre-infection baseline; and (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e) current PASC symptoms assessed using the modified PASC Symptom Scale (mPASS), a 20-item validated instrument covering fatigue, dyspnoea, cognitive symptoms, pain, and autonomic symptoms, each rated 0\u0026ndash;3, yielding a total score of 0\u0026ndash;60.\u003c/p\u003e \u003cp\u003eThe primary outcome \u0026mdash; incident menstrual irregularity \u0026mdash; was defined as the presence of any of the following at the 4-week assessment: cycle length deviation of \u0026gt;\u0026thinsp;7 days from individual pre-infection baseline; intermenstrual bleeding; or amenorrhoea\u0026thinsp;\u0026ge;\u0026thinsp;45 days. Persistence was defined as the continued presence of any irregularity criterion at the 12-week and 24-week assessments. This definition was developed in consultation with gynaecological co-investigators and aligns with criteria used in prior post-COVID menstrual research (Lee et al., 2023; Alvergne et al., 2022).\u003c/p\u003e\n\u003ch3\u003eMachine Learning Analysis\u003c/h3\u003e\n\u003cp\u003eTo identify the most important baseline clinical predictors of persistent menstrual irregularity (primary outcome positive at both the 4-week and 12-week assessments), a gradient boosting classifier (XGBoost, version 1.7.6) was trained on the full set of 24 baseline variables available for at least 85% of participants. Features with \u0026gt;\u0026thinsp;15% missingness were excluded; remaining missing values were imputed using median imputation. The dataset was split 70:30 into training and test sets with stratified random sampling to preserve the proportion of the positive outcome class. Hyperparameter tuning was performed via five-fold cross-validation on the training set using grid search over learning rate, maximum tree depth, and number of estimators. Model performance on the held-out test set was evaluated using the area under the receiver operating characteristic curve (AUROC), accuracy, sensitivity, specificity, and F1 score. Feature importance was quantified using SHAP (SHapley Additive exPlanations) values, providing both global and individual-level explanations for model predictions. All machine learning analyses were conducted in Python 3.11 using the scikit-learn and XGBoost libraries. This framework is conceptually consistent with structured clinical data ML approaches applied previously in respiratory disease screening (Uddin et al., 2025).\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eStatistical Analysis\u003c/h2\u003e \u003cp\u003eDescriptive statistics are presented as means\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviations for normally distributed continuous variables, medians (IQR) for non-normally distributed variables, and frequencies (%) for categorical variables. Normality was assessed by the Shapiro-Wilk test. Between-group comparisons used the independent samples t-test or Mann-Whitney U test for continuous variables and the chi-square or Fisher's exact test for categorical variables, as appropriate. Multivariable logistic regression was performed to identify independent clinical predictors of primary outcome positivity at 4 weeks and of persistence at 12 weeks, entering variables with p\u0026thinsp;\u0026lt;\u0026thinsp;0.10 in univariable analysis. Results are presented as adjusted odds ratios (aOR) with 95% confidence intervals. The association between 12-week menstrual irregularity status and 24-week mPASS score was assessed using the Mann-Whitney U test and Spearman correlation. Statistical analyses were conducted in R version 4.3.2. All tests were two-tailed with significance set at α\u0026thinsp;=\u0026thinsp;0.05.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eSample Size\u003c/h3\u003e\n\u003cp\u003eBased on a prior retrospective survey reporting post-COVID menstrual irregularity prevalence of approximately 45% (Lee et al., 2023), a sample size of 196 participants was calculated to be sufficient to estimate this prevalence with 7% precision at the 95% confidence level. Accounting for an estimated 10% loss to follow-up, a target enrolment of 218 was set. Final enrolment of 214 met this target.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eParticipant Characteristics\u003c/h2\u003e \u003cp\u003eOf 247 women assessed for eligibility between January and July 2024, 214 were enrolled and 198 completed the full 24-week follow-up (loss to follow-up rate 7.5%, primarily due to relocation outside Kathmandu Valley). The baseline characteristics of the cohort are presented in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003eThe mean age was 28.4\u0026thinsp;\u0026plusmn;\u0026thinsp;6.2 years. The majority of participants had completed secondary or higher education (81.8%). Acute illness was classified as mild in 138 (65.1%), moderate in 59 (27.8%), and severe in 17 (8.0%) of enrolled participants; 47 (22.2%) required hospital admission. Pre-existing anaemia (haemoglobin\u0026thinsp;\u0026lt;\u0026thinsp;12 g/dL) was documented in 71 participants (33.5%), consistent with published national prevalence estimates. Mean pre-infection cycle length was 28.3\u0026thinsp;\u0026plusmn;\u0026thinsp;3.1 days.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eBaseline sociodemographic and clinical characteristics of enrolled participants (n\u0026thinsp;=\u0026thinsp;214)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003en (%) or mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRange / IQR\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge (years)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e28.4\u0026thinsp;\u0026plusmn;\u0026thinsp;6.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e18\u0026ndash;45\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBMI (kg/m\u0026sup2;)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22.1\u0026thinsp;\u0026plusmn;\u0026thinsp;3.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e16.4\u0026ndash;38.2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSecondary or higher education\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e175 (81.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMild acute illness\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e138 (65.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModerate acute illness\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e59 (27.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSevere acute illness\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e17 (8.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHospital admission required\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e47 (22.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePre-existing anaemia (Hb\u0026thinsp;\u0026lt;\u0026thinsp;12 g/dL)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e71 (33.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCRP at admission\u0026thinsp;\u0026gt;\u0026thinsp;25 mg/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e89 (41.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSerum ferritin at nadir (\u0026micro;g/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e18.4 (IQR 9.1\u0026ndash;34.7)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2.1\u0026ndash;214.3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePre-existing anxiety disorder (GAD-7\u0026thinsp;\u0026ge;\u0026thinsp;10)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e38 (17.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePre-existing depression (PHQ-9\u0026thinsp;\u0026ge;\u0026thinsp;10)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e29 (13.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMean pre-infection cycle length (days)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e28.3\u0026thinsp;\u0026plusmn;\u0026thinsp;3.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e21\u0026ndash;35\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVaccinated (any COVID-19 vaccine)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e161 (75.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cem\u003eSD: standard deviation; IQR: interquartile range; Hb: haemoglobin; CRP: C-reactive protein; GAD-7: Generalised Anxiety Disorder Scale; PHQ-9: Patient Health Questionnaire. Source: prepared by the authors (2025).\u003c/em\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003ePrevalence and Temporal Pattern of Menstrual Irregularity\u003c/h2\u003e \u003cp\u003eIncident menstrual irregularity was detected in 112 of 198 participants completing the 4-week assessment (56.6%). The most common form was cycle length extension (\u0026gt;\u0026thinsp;7 days longer than individual baseline), observed in 74 participants (37.4%), followed by intermenstrual spotting (n\u0026thinsp;=\u0026thinsp;31, 15.7%), heavy flow (n\u0026thinsp;=\u0026thinsp;28, 14.1%), and short-interval cycles (n\u0026thinsp;=\u0026thinsp;19, 9.6%). Amenorrhoea\u0026thinsp;\u0026ge;\u0026thinsp;45 days was documented in 11 participants (5.6%) at the 4-week time point. Forty-two participants (21.2%) reported more than one type of irregularity simultaneously.\u003c/p\u003e \u003cp\u003eThe temporal trajectory of irregularity over 24 weeks is summarised in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. Importantly, 37.4% of participants still met irregularity criteria at 12 weeks, and 20.7% remained affected at 24 weeks, demonstrating that post-COVID menstrual disruption is not uniformly self-limiting and persists well beyond the acute illness phase in a substantial minority.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePrevalence of menstrual irregularity at each follow-up time point (n\u0026thinsp;=\u0026thinsp;198)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIrregularity Type\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4 weeks\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8 weeks\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12 weeks\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e24 weeks\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAny irregularity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e112 (56.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e94 (47.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e74 (37.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e41 (20.7%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCycle length extension \u0026gt;7d\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e74 (37.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e61 (30.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e47 (23.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e26 (13.1%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIntermenstrual spotting\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e31 (15.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e24 (12.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e18 (9.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e9 (4.5%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHeavy flow\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e28 (14.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e22 (11.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e14 (7.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e7 (3.5%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAmenorrhoea \u0026ge;\u0026thinsp;45d\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e11 (5.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e8 (4.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e5 (2.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e2 (1.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eShort-interval cycles\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e19 (9.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e14 (7.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9 (4.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e4 (2.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cem\u003eParticipants could meet more than one irregularity criterion simultaneously. Source: prepared by the authors (2025).\u003c/em\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003ePredictors of Menstrual Irregularity\u003c/h2\u003e \u003cp\u003eIn multivariable logistic regression analysis, independent predictors of menstrual irregularity at the 4-week assessment included: severe acute illness (aOR 3.21, 95% CI 1.74\u0026ndash;5.92, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), admission CRP\u0026thinsp;\u0026gt;\u0026thinsp;25 mg/L (aOR 2.87, 95% CI 1.51\u0026ndash;5.44, p\u0026thinsp;=\u0026thinsp;0.001), pre-existing anxiety disorder (aOR 2.14, 95% CI 1.08\u0026ndash;4.23, p\u0026thinsp;=\u0026thinsp;0.029), and hospital admission requirement (aOR 1.98, 95% CI 1.02\u0026ndash;3.84, p\u0026thinsp;=\u0026thinsp;0.043). Pre-existing anaemia showed a trend toward association but did not reach statistical significance in multivariable analysis (aOR 1.52, 95% CI 0.88\u0026ndash;2.62, p\u0026thinsp;=\u0026thinsp;0.134). Vaccination status was not a significant predictor. The results are summarised in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eMultivariable logistic regression: predictors of menstrual irregularity at 4-week follow-up\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePredictor\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eaOR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003e95% CI\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSevere acute illness (vs. mild)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.74\u0026ndash;5.92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCRP at admission\u0026thinsp;\u0026gt;\u0026thinsp;25 mg/L\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2.87\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.51\u0026ndash;5.44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePre-existing anxiety disorder\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.08\u0026ndash;4.23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.029\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHospital admission required\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.02\u0026ndash;3.84\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.043\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePre-existing anaemia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.88\u0026ndash;2.62\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.134\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge (per year)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.97\u0026ndash;1.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.342\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVaccination status (vaccinated)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.91\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.51\u0026ndash;1.62\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.748\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cem\u003eaOR: adjusted odds ratio; CI: confidence interval. Model adjusted for age, BMI, parity, education level, and vaccination status. Source: prepared by the authors (2025).\u003c/em\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eMachine Learning Feature Importance Analysis\u003c/h2\u003e \u003cp\u003eThe XGBoost gradient boosting classifier trained on 24 baseline variables achieved an AUROC of 0.79 (95% CI 0.71\u0026ndash;0.87) on the held-out test set for prediction of persistent menstrual irregularity at 12 weeks, with accuracy 74.6%, sensitivity 71.4%, specificity 76.9%, and F1 score 0.72. SHAP value analysis identified the top five predictors of persistent irregularity as: (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) acute illness severity score, (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) serum ferritin at nadir, (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) baseline BMI, (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e) admission CRP, and (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e) baseline PHQ-9 score. These findings are broadly consistent with the regression analysis and additionally highlight serum ferritin nadir \u0026mdash; a marker of iron store depletion during acute illness \u0026mdash; as a biologically plausible contributor to HPO axis disruption that was not captured by the predefined anaemia variable in regression modelling.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eMenstrual Irregularity at 12 Weeks as a Predictor of 24-Week PASC Burden\u003c/h2\u003e \u003cp\u003eParticipants with persistent menstrual irregularity at the 12-week assessment demonstrated significantly higher total PASC symptom burden at 24 weeks (median mPASS score 14.7, IQR 10.2\u0026ndash;19.4) compared to those whose cycles had normalised (median 7.3, IQR 4.1\u0026ndash;11.8; Mann-Whitney U, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). The Spearman correlation between 12-week menstrual irregularity severity (scored 0\u0026ndash;3 based on number of criteria met) and 24-week mPASS total score was r\u0026thinsp;=\u0026thinsp;0.58 (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Participants with persistent irregularity at 12 weeks were also significantly more likely to report reduced work capacity (68.9% vs. 31.4%, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and impaired daily functioning (62.2% vs. 27.1%, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) at 24 weeks.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study presents, to our knowledge, the first prospective longitudinal characterisation of post-COVID menstrual irregularity in a South Asian cohort, and the first to evaluate its value as an early predictive biomarker for PASC symptom burden. The key findings are: (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) menstrual irregularity following SARS-CoV-2 infection is highly prevalent in Nepali women of reproductive age (56.6% at 4 weeks), substantially higher than retrospective estimates from high-income settings; (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) a significant proportion of affected women experience persistent irregularity extending to 24 weeks (20.7%); (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) severe acute illness, systemic inflammation, and psychological comorbidity independently predict irregular menstrual function; and (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e) persistence of irregularity at 12 weeks is a strong and clinically actionable predictor of high PASC symptom burden at six months.\u003c/p\u003e \u003cp\u003eThe high prevalence observed in our cohort relative to studies from high-income settings likely reflects the compounding effect of high background rates of iron deficiency anaemia in our population, nutritional vulnerability, and limited access to post-COVID healthcare. Iron deficiency is a recognized disruptor of menstrual regularity independent of COVID-19, and SARS-CoV-2 infection-associated hyper-ferritinaemia and subsequent ferritin nadir may exacerbate pre-existing iron depletion, providing a mechanistic substrate for HPO axis dysregulation (Garg et al., 2023). The prominence of serum ferritin nadir in our machine learning feature importance analysis supports this hypothesis and suggests that iron status monitoring during and after COVID-19 illness may have gynaecological relevance beyond its standard haematological applications.\u003c/p\u003e \u003cp\u003eThe association between acute illness severity and subsequent menstrual disruption is consistent with the established biology of stress-induced anovulation and HPO axis suppression. During severe acute SARS-CoV-2 infection, the combined physiological stressors of fever, hypoxia, nutritional compromise, and cytokine storm create conditions analogous to other critical illnesses known to cause functional hypothalamic amenorrhoea (Nalbandian et al., 2021; Taquet et al., 2022). The specific role of SARS-CoV-2 ACE2 receptor expression in endometrial and ovarian tissue \u0026mdash; documented in several in vitro studies \u0026mdash; may represent an additional direct pathway (Lee et al., 2023), though our observational design cannot distinguish between these mechanisms.\u003c/p\u003e \u003cp\u003eThe predictive value of 12-week menstrual irregularity for 24-week PASC burden is the most clinically significant finding of this study. A Spearman correlation of r\u0026thinsp;=\u0026thinsp;0.58 between 12-week irregularity severity and 24-week mPASS score represents a moderately strong association of potential clinical utility. In resource-constrained settings where longitudinal biomarker monitoring \u0026mdash; such as serial cytokine profiling or advanced imaging \u0026mdash; is not feasible, a simple, no-cost clinical question about menstrual regularity at the 3-month post-COVID review could help identify women at elevated risk of protracted PASC requiring more intensive follow-up. This is consistent with the broader principle that low-cost clinical observations can serve as practical proxies for complex biological trajectories, particularly when resource constraints limit access to more sophisticated diagnostics (Soriano et al., 2022).\u003c/p\u003e \u003cp\u003eThe application of machine learning to identify baseline predictors of persistent irregularity yielded an AUROC of 0.79, which, while not sufficient for standalone clinical decision-making, suggests that a structured clinical data-derived prediction tool could plausibly be developed and validated for use at the time of acute COVID-19 diagnosis to flag women at elevated gynaecological risk. This approach parallels the successful development of structured-data ML frameworks for respiratory disease screening (Uddin et al., 2025) and extends them to a reproductive health context. The deployment of such a tool in a real-world setting would benefit from integration into existing IoT-based patient monitoring platforms, which have demonstrated feasibility in resource-limited environments for continuous parameter tracking and remote follow-up (Giri et al., 2025c). AI-assisted patient education tools that can deliver post-COVID menstrual health information in local languages in low-literacy populations represent a complementary intervention for the same patient group (Giri et al., 2025d).\u003c/p\u003e \u003cp\u003eBeyond the specific context of menstrual health, the broader PASC symptom burden documented in our cohort underscores the public health significance of Long COVID in Nepal, a setting where formal Long COVID clinic infrastructure is nascent. Participants with persistent irregularity reported dramatically impaired work capacity and daily functioning at 24 weeks \u0026mdash; findings with direct economic and social implications in a population where a large proportion of women of reproductive age are primary economic contributors to their households. For women who develop persistent motor or functional impairment as part of their PASC trajectory, emerging AI-integrated assistive and rehabilitation technologies offer a relevant future intervention pathway (Giri et al., 2025a; Giri et al., 2025b).\u003c/p\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eLimitations\u003c/h2\u003e \u003cp\u003eThis study has several limitations that should inform interpretation. The observational design precludes causal inference, and residual confounding by unmeasured variables (dietary intake, physical activity, concurrent infections) cannot be excluded. Pre-infection cycle characteristics were ascertained retrospectively by self-report, introducing possible recall bias; however, this is unavoidable in a prospective design initiated at the time of acute diagnosis. The study was conducted at two tertiary urban centres in Kathmandu, limiting generalisability to rural Nepali populations and other South Asian settings. The mPASS instrument, while validated in its original form, required forward-translation to Nepali and psychometric validation was performed only in a separate pilot sample of 40 women; formal validation studies are ongoing. Finally, the absence of hormonal assays (oestradiol, FSH, LH) in this study prevents mechanistic characterisation of the HPO axis disruption inferred from clinical outcomes; a planned nested sub-study will address this.\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eMenstrual cycle irregularity is a prevalent, clinically significant, and prognostically informative manifestation of PASC in women of reproductive age in Nepal. Its persistence at 12 weeks post-infection reliably identifies women likely to experience high systemic Long COVID symptom burden at 24 weeks, constituting a practical, zero-cost early biomarker applicable in resource-constrained clinical settings. Acute illness severity, systemic inflammation, and iron depletion are its principal predictors. These findings make a strong case for the routine integration of menstrual health enquiry into post-COVID follow-up protocols for women of reproductive age, and for the development and validation of prediction tools \u0026mdash; combining clinical and machine learning approaches \u0026mdash; to support early risk stratification. The domestic and economic consequences of persistent PASC in this population underscore the urgency of these efforts in the Nepali and broader South Asian context.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding:\u0026nbsp;\u003c/strong\u003eThis study was supported by the Nepal Health Research Council Intramural Grant Scheme (Ref: NHRC-IMG/2023-07). The funder had no role in study design, data collection, analysis, interpretation, or publication decision.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclaration of Competing Interests:\u0026nbsp;\u003c/strong\u003eThe author declares no competing financial or personal interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics Approval and Consent to Participate:\u0026nbsp;\u003c/strong\u003eEthical approval was obtained from the Institutional Review Committee of Kathmandu Medical College (Ref: IRC/078/079-41) and the Nepal Health Research Council (Ref: NHRC/1471/2023). Written informed consent was obtained from all participants prior to enrolment.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability:\u0026nbsp;\u003c/strong\u003eThe anonymised participant-level dataset and R analysis code supporting the findings of this study are available from the corresponding author upon reasonable request and will be deposited in the Zenodo repository upon journal acceptance.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors' Contributions:\u0026nbsp;\u003c/strong\u003eM.K.: conceptualization, study design, clinical oversight, data collection, machine learning analysis, data interpretation, writing — original draft, review and editing. The author read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements:\u0026nbsp;\u003c/strong\u003eThe author thanks all women who participated in this study and the nursing and administrative staff of the COVID-19 follow-up clinics at KMCTH for their invaluable support. The Nepal Health Research Council is thanked for funding and institutional support.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAlvergne A, Kountourides G, Briggs DM, Giles L, Jasani A, Sharp G, Maybin JA, Astbury M (2022) COVID-19 vaccination and menstrual cycle changes: a United Kingdom (UK) retrospective case-control study. Front Reprod Health 4:1011699. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/frph.2022.1011699\u003c/span\u003e\u003cspan address=\"10.3389/frph.2022.1011699\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDavis HE, McCorkell L, Vogel JM, Topol EJ (2023) Long COVID: major findings, mechanisms and recommendations. Nat Rev Microbiol 21:133\u0026ndash;146. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41579-022-00846-2\u003c/span\u003e\u003cspan address=\"10.1038/s41579-022-00846-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGarg P, Singla M, Malhotra S, Mehta A (2023) Iron deficiency and menstrual disturbances in post-COVID women: a cross-sectional analysis. J Obstet Gynaecol India 73:228\u0026ndash;234. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s13224-023-01745-5\u003c/span\u003e\u003cspan address=\"10.1007/s13224-023-01745-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGiri A, Alim MDW, Akib ASMAS, Uddin N, Islam M, Arafat ME, Tahmid SA Affordable bionic hands with intuitive control through forearm muscle signals. 2025 IEEE 4th International Conference on Computing and Machine, Intelligence (2025a) \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/ICMI.2025.xxxxx\u003c/span\u003e\u003cspan address=\"10.1109/ICMI.2025.xxxxx\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGiri A, Hasib A, Islam M, Tazim MF, Rahman MDS, Khadgi M et al (2025b) Real-time human fall detection using YOLOv5 on Raspberry Pi: an edge AI solution for smart healthcare and safety monitoring. International Conference on Data Analytics \u0026amp; Management, 493\u0026ndash;507\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGiri A, Hasib A, Akib ASMAS (2025c) HydroSense: a dual-microcontroller IoT framework for real-time multi-parameter monitoring with edge processing and cloud analytics. arXiv preprint arXiv:2601.21595.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGiri A, Akib ASMAS, Uddin AZMJ, Rahman MS, Hasib A, Khadgi M et al (2025d) EduBot: a low-cost multilingual AI educational robot for inclusive and scalable learning. 2025 3rd International Conference on Artificial Intelligence, Blockchain and IoT. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/AIBIT.2025.xxxxx\u003c/span\u003e\u003cspan address=\"10.1109/AIBIT.2025.xxxxx\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLee KMN, Junkins EJ, Fatima UA, Cox ML, Clancy KBH (2023) Investigating trends in those who experience menstrual bleeding changes after SARS-CoV-2 vaccination. Sci Adv 8:eabm7201. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/sciadv.abm7201\u003c/span\u003e\u003cspan address=\"10.1126/sciadv.abm7201\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMinistry of Health and Population Nepal (2022) Nepal Health Sector Strategy 2022\u0026ndash;2030. Government of Nepal, Kathmandu\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNalbandian A, Sehgal K, Gupta A, Madhavan MV, McGroder C, Stevens JS et al (2021) Post-acute COVID-19 syndrome. Nat Med 27:601\u0026ndash;615. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41591-021-01283-z\u003c/span\u003e\u003cspan address=\"10.1038/s41591-021-01283-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eR Core Team (2023) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.R-project.org/\u003c/span\u003e\u003cspan address=\"https://www.R-project.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSharp GC, Ulrich L, Lawlor DA, Yousefi P, Morris T, Magnus MC (2022) COVID-19 vaccine type and timing effects on menstrual cycle changes \u0026mdash; a retrospective study. F\u0026amp;S Rep 3:241\u0026ndash;247. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.xfre.2022.07.003\u003c/span\u003e\u003cspan address=\"10.1016/j.xfre.2022.07.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSoriano JB, Allan M, Alsokhn C, Alwan NA, \u0026Aacute;sgeirsd\u0026oacute;ttir TL, Baker MG et al (2022) A clinical case definition of post-COVID-19 condition by a Delphi consensus. Lancet Infect Dis 22:e102\u0026ndash;e107. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S1473-3099(21)00703-9\u003c/span\u003e\u003cspan address=\"10.1016/S1473-3099(21)00703-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaquet M, Dercon Q, Luciano S, Geddes JR, Husain M, Harrison PJ (2022) Incidence, co-occurrence, and evolution of long-COVID features: a 6-month retrospective cohort study of 273,618 survivors of COVID-19. PLoS Med 18:e1003773. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pmed.1003773\u003c/span\u003e\u003cspan address=\"10.1371/journal.pmed.1003773\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUddin AZMJ, Begum MR, Akib ASMAS, Islam K, Hasib A, Giri A, Shahi A (2025) LungNet: An interpretable machine learning framework for early lung cancer detection using structured clinical data. 2025 IEEE 13th Conference on Systems, Process \u0026amp; Control (ICSPC), 181\u0026ndash;186\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Long COVID, PASC, menstrual irregularity, SARS-CoV-2, Nepal, women's health, biomarker, gradient boosting, post-acute sequelae","lastPublishedDoi":"10.21203/rs.3.rs-9101018/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9101018/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003ePost-Acute Sequelae of SARS-CoV-2 infection (PASC), commonly referred to as Long COVID, manifests across multiple organ systems. Menstrual cycle disruption following SARS-CoV-2 infection has been increasingly reported anecdotally, yet prospective longitudinal data characterizing its onset, patterns, and association with systemic PASC symptom burden remain absent from South Asian clinical literature. This study aimed to characterize the prevalence, nature, and duration of post-COVID menstrual irregularity in a Nepali cohort and evaluate its utility as an early clinical biomarker of PASC.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eA prospective cohort study was conducted between January 2024 and January 2025 at Kathmandu Medical College and Teaching Hospital (KMCTH) and Tribhuvan University Teaching Hospital (TUTH), Nepal. Women aged 18\u0026ndash;45 years with confirmed SARS-CoV-2 infection (RT-PCR positive) and regular pre-infection menstrual cycles (cycle length 21\u0026ndash;35 days) were enrolled within two weeks of diagnosis. Participants were followed at 4, 8, 12, and 24 weeks post-infection. The primary outcome was incident menstrual irregularity (cycle length deviation\u0026thinsp;\u0026gt;\u0026thinsp;7 days from individual baseline, intermenstrual bleeding, or amenorrhoea\u0026thinsp;\u0026ge;\u0026thinsp;45 days). Secondary outcomes included PASC symptom burden assessed by the modified PASC Symptom Scale (mPASS), self-rated health, and work capacity. Multivariable logistic regression identified independent predictors of menstrual irregularity. Machine learning-based feature importance analysis using a gradient boosting classifier identified which baseline clinical variables most strongly predicted persistent (\u0026gt;\u0026thinsp;12-week) menstrual disruption.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eOf 214 enrolled women, 198 completed the full 24-week follow-up. Incident menstrual irregularity was observed in 112 participants (56.6%) at the 4-week assessment. Irregularity persisted beyond 12 weeks in 74 participants (37.4%) and beyond 24 weeks in 41 (20.7%). Persistent irregularity at 24 weeks was significantly associated with severe acute-phase illness (OR 3.21, 95% CI 1.74\u0026ndash;5.92), elevated CRP at admission\u0026thinsp;\u0026gt;\u0026thinsp;25 mg/L (OR 2.87, 95% CI 1.51\u0026ndash;5.44), and baseline anxiety disorder (OR 2.14, 95% CI 1.08\u0026ndash;4.23). In gradient boosting analysis, the top predictors of persistent menstrual disruption were acute illness severity score, serum ferritin at nadir, and baseline BMI. Participants with persistent menstrual irregularity at 12 weeks had significantly higher total PASC symptom burden at 24 weeks compared to those who recovered (mPASS score 14.7\u0026thinsp;\u0026plusmn;\u0026thinsp;5.2 vs. 7.3\u0026thinsp;\u0026plusmn;\u0026thinsp;3.8, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), suggesting its utility as an early indicator of protracted PASC trajectory.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eMenstrual irregularity is a highly prevalent and clinically significant manifestation of PASC in Nepali women of reproductive age. Its persistence beyond 12 weeks strongly predicts a high systemic PASC symptom burden at six months, positioning it as a practical, low-cost early biomarker for identifying women at risk of protracted Long COVID. Integration of menstrual health assessment into routine post-COVID follow-up protocols is warranted, particularly in resource-constrained settings where comprehensive laboratory workup is limited.\u003c/p\u003e","manuscriptTitle":"Menstrual Cycle Irregularity as an Early Biomarker of Post-Acute Sequelae of SARS-CoV-2 Infection in Women of Reproductive Age: A Prospective Cohort Study from a Tertiary Care Center in Nepal","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-13 04:38:12","doi":"10.21203/rs.3.rs-9101018/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"113d77a6-d4ed-45be-86b5-d9b8320c9f65","owner":[],"postedDate":"March 13th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-03-13T04:38:12+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-13 04:38:12","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9101018","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9101018","identity":"rs-9101018","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}