key: cord-0859914-y54hkuuq
authors: Hoong, Caroline Wei Shan; Amin, Muhammad Nakib Monjur E; Tan, Teck Choon; Lee, Jer En
title: Viral arthralgia a new manifestation of COVID-19 infection? A cohort study of COVID-19-associated musculoskeletal symptoms
date: 2021-01-18
journal: Int J Infect Dis
DOI: 10.1016/j.ijid.2021.01.031
sha: 820dc9b19c45ea5ecfc396d4578888654e3783b4
doc_id: 859914
cord_uid: y54hkuuq

OBJECTIVES: Musculoskeletal symptoms are often not recognised as a prominent feature of COVID-19 infection. We hypothesised that viral arthralgia is an uncommon but distinct manifestation of COVID-19 infection. In addition, we aimed to characterise the other musculoskeletal presentations of COVID-19 infection and study their prognostic implications. METHODS: Patients hospitalised with COVID-19 infection were divided into 2 groups: those with and without musculoskeletal symptoms. Those with musculoskeletal symptoms were subdivided according to 4 patterns of musculoskeletal involvement: myalgia, arthralgia, backache and generalised bodyache. Using binary regression logistic analysis, we compared the risk of developing a viral pneumonia in patients with and without musculoskeletal complaints. RESULTS: Of 294 hospitalised patients with COVID-19, 88 (30%) reported musculoskeletal complaints. Among these 88 patients, 37.5% had myalgia, 5.7% arthralgia, 6.8% new-onset backache and 50% generalized bodyache. The presence of musculoskeletal complaints was not associated with the risk of developing viral pneumonia (6.8% versus 9.7%, OR: 0.68, 95%CI: 0.26-1.76, p = 0.426). COVID-19 arthralgia was often more severe and had variable onset, while generalised bodyache and myalgia were milder and coincided with the occurrence of fever or respiratory symptoms. CONCLUSION: Viral arthralgia is novel clinical manifestation of COVID-19, not typical of a viral prodrome nor a reactive arthropathy. While musculoskeletal symptoms were not associated with developing a pneumonia, clinicians should be aware of its variable onset to avoid missing a diagnosis of COVID-19, particularly when respiratory symptoms are absent at the time of presentation.

Singapore was one of the first countries to report its index case of COVID-19 on 23 January 2020 (Chotirmall et al., 2020) . Since then, it has been one of the most successful in limiting community spread by extensive testing, meticulous contact tracing, active screening of at-risk groups, pre-emptive hospitalisation and mandatory mask-wearing (Chotirmall et al., 2020) . As a result, hospitals have remained well-equipped and mortality rates from COVID-19 remains one of the lowest globally (Worldometers, 2020) . The vast majority of patients with COVID-19 admitted to hospitals in Singapore remain clinically well; instead, the authors have noticed musculoskeletal symptoms to be a prominent complaint among them. Particularly interesting is a small subgroup of patients who present with a viral arthralgia, which may be an uncommon but currently under-recognised new clinical entity in COVID-19 , Zheng et al., 2020 .

COVID-19 is a global pandemic of predominantly respiratory tract but also multi-organ involvement. J o u r n a l P r e -p r o o f Viral arthralgia in COVID-19 has been described in isolated case reports (Parisi et al., 2020 , Liew et al., 2020 , Joob and Wiwanitkit, 2020 ), but their significance as a new clinical entity has not been fully appreciated.

Interestingly, we noticed that they have been reported to occur independently of respiratory symptoms in COVID-19. One case in Italy occurred as a reactive arthritis 25 days after the onset of acute respiratory symptoms (Parisi et al., 2020) . A second case of reactive arthritis diagnosed with COVID-19 from active screening was completely asymptomatic for respiratory symptoms (Liew et al., 2020) , and a third case in Thailand presented with fever and arthralgia which pre-dated the onset of respiratory symptoms (Joob and Wiwanitkit, 2020), leading to a delayed diagnosis of COVID-19 infection. Without a high clinical index of suspicion and adequate follow-up, the diagnosis may be easily missed if an overemphasis on respiratory symptoms is placed.

We performed a descriptive cohort study of hospitalised patients with COVID-19 infection diagnosed by reverse transcription polymerase chain reaction (RT-PCR) admitted to the general wards of Woodlands Health Campus in Singapore. The aim of this study was to characterise the prevalence, onset and nature of musculoskeletal complaints in COVID-19 patients and their clinical significance. We hypothesised that viral arthralgia is an uncommon but distinct manifestation of COVID-19 infection, which could present days before or after the onset of respiratory symptoms. In addition, we hypothesised that the presence of musculoskeletal symptoms was not associated with the development of a viral pneumonia.

A single-centre retrospective cohort study was conducted of patients to Woodlands Health Campus Singapore between 1 April to 31 May 2020. Subjects were included if they were age >18, of general ward status at the time of admission and diagnosed with COVID-19 via RT-PCR from nasopharyngeal and throat J o u r n a l P r e -p r o o f swab specimen. This study was approved by the Institution Ethics Board (DSRB 2020/00765) and strict patient confidentiality was maintained.

Exclusion criteria included pregnant patients, those with severe debilitating end-stage disease on palliative care, those with missing documentation of the onset and nature of presenting symptoms, missing information on the outcome of developing a pneumonia, or those who were still hospitalised at the end of data collection because the endpoint of development of pneumonia was unknown. At the time of completion of data collection on 31 July 2020, all patients had been discharged. No subject was excluded based on any of the criteria above.

Musculoskeletal complaints were considered to be related to COVID-19 infection if they were of new onset within 2 weeks prior or 1 month after the diagnosis of COVID-19 infection. Pain localizing to the chest, abdomen or headaches were not considered musculoskeletal in nature. Subdivision into groups of myalgia, arthralgia, backache, or generalised bodyaches were determined by localisation to the muscle, joint(s), back or unspecified respectively, as documented in the detailed history in the case notes. If there were two or more types of musculoskeletal symptoms, patients were classified according to their predominant complaint.

Data was obtained from electronic health records. We collected information on patient demographics, comorbidities, history of presenting complaint including pain score, duration of pain, onset and distribution, development of fever or respiratory symptoms, clinical examination, use of analgesia, length of stay, as well as COVID-19-related biochemical characteristics on admission, including C-reactive protein (CRP), lactate dehydrogenase (LDH), total white cell count (TW), lymphocyte, neutrophil and platelet count and dengue serology for the presence of co-infection. Pain score was taken to be the maximum recorded on the first 72 J o u r n a l P r e -p r o o f hours of admission, as assessed by visual analogue scale from 0 to 10. Daily case notes were inspected till day of discharge, particularly looking for new-onset musculoskeletal symptoms, clinical and radiological evidence of pneumonia, that developed during the admission. Two doctors (CWSH and MNMEA) extracted all the pre-defined variables, and the data was cross-checked for integrity before analysis. Where there were different interpretations, these were resolved by mutual consensus and adjudication by the principal investigator (LJE).

Our study was primarily descriptive to highlight a new clinical entity. Outcomes analysed were the development of pneumonia as determined radiologically and clinically, as documented in case sheets, requirement for supplemental oxygen, intensive care unit (ICU) care and mortality. The requirement for supplemental oxygen was defined as finger oxygen saturation <93% on room air, or arterial blood oxygen partial pressure (PaO2) / oxygen concentration (FiO2) < 300mmHg.

Statistics were performed using SPSS Statistics for Windows, Version 26.0 (IBM Corp, Armonk, New York).

Continuous variables were presented as mean ± SD or median (interquartile range), and categorical data were shown as n (%). Clinical characteristics were compared between patients with and without musculoskeletal symptoms, and between patients in the different musculoskeletal groups using the Mann-Whitney U test, Fisher's exact test and one-way ANOVA as appropriate. Binary logistic regression analysis was used to identify factors associated with the development of pneumonia. Considering that the number developing pneumonia was small, our model included the limited covariates of age, diabetes, hypertension and presence of musculoskeletal symptoms, fever, and respiratory symptoms to avoid overfitting. A twotailed p-value of <0.05 was considered statistically significant.

A total of 294 patients were hospitalised for COVID-19 infection ( Table 1 ). The median age was 36 (IQ, interquartile range 30-45) years and all were male. The young age and overwhelming predominance of male patients among COVID-19 infections was due to the disease outbreak in foreign worker dormitories and a low community transmission rate in Singapore (Ministry of Health Singapore, 2020). Musculoskeletal complaints were reported in 30% of the cohort. Of the 88 patients with musculoskeletal symptoms, 37.5% had myalgia, 5.7% arthralgia, 6.8% new-onset backache and 50% generalized bodyache. Patients with arthralgia were non-significantly older than those without arthralgia (median age 44 versus 36 years, p=0.078). Only 1 patient in this cohort had a history of rheumatological disease (gout), but he did not develop musculoskeletal symptoms during this hospitalisation.

Comparing patients with and without musculoskeletal symptoms, age and comorbidities were similar in both groups (Table 1) . Patients with musculoskeletal symptoms had a higher prevalence of fever (p<0.01) and a higher CRP level (p<0.01). Respiratory symptoms were similar in both groups without and without musculoskeletal symptoms. LDH, platelet count, TW, lymphocyte, neutrophil count and neutrophillymphocyte ratio were not significantly different. CRP was significantly higher in patients with compared to without musculoskeletal symptoms, but levels were similar among the 4 groups of patients with various musculoskeletal symptoms. LDH, a marker of inflammation as well as tissue damage, was similar across all 4 musculoskeletal symptom groups.

Patients with myalgia, backache or generalized bodyache experienced only mild pain on admission, with a median maximum pain score of 0 (IQ 0, 2), 0 (IQ 0, 1), 0 (IQ 0, 0) ( Table 2 ) upon 10. In contrast, those with arthralgia reported a higher median pain score of 2 upon 10 (IQ 0, 5) on admission and required more analgesia than all other groups. Majority of patients in all groups experienced fever and/or respiratory J o u r n a l P r e -p r o o f symptoms. 25% of subjects having musculoskeletal symptoms experienced fever without respiratory symptoms. While the onset of the generalized bodyache or myalgia coincided with the onset of fever or respiratory symptoms, arthralgia could occur before, during or after the onset of fever or respiratory symptoms (range between 12 days before to 6 days later), and backache occurred more than 2 (range 0-13) days later. Musculoskeletal symptoms were more persistent in the arthralgia group, lasting a median of 9 (IQ 7,18) days, as compared to 3 (IQ 2,4) days, 3 (IQ 2,4) days and 1.5 (IQ 1,5) days in the other groups with generalized bodyache, myalgia and backache respectively.

The distribution of joint involvement was unique in each of the 5 patients who had a viral arthralgia (Table   3) . Two patients had an oligoarthropathy, 1 had a monoarthropathy affecting the shoulder and 2 had a polyarthropathy. Three patients had only large joint involvement in the knees, shoulders and elbows, 1 only small joint involvement in the proximal interphalangeal joint (PIPJ) and metacarpophalangeal joints (MCPJ) of the hand, and one had both large and small joint involvement. The affected joints were symmetrically distributed in 3 patients and unilateral in 2 patients. None of the 5 patients had joint swelling or effusion or evidence of synovial inflammation on clinical examination, nor did they have an associated rash, enteritis, conjunctivitis or urethritis. In addition, one patient (Patient 3) had evidence of tendonitis over the extensor tendons of bilateral elbows. Patient 1 was tested negative for Gonococcal and Chlamydia trachomatis PCR from urethral swab, and was negative for rheumatoid factor (data not shown). None of the five patients with arthralgia had a past history of osteoarthritis, gout or an alternate rheumatic disease. 80% of patients with arthralgia experienced persistent pain lasting more than 1 week, and patient 1 experienced a protracted polyarthralgia lasting 29 days (Table 3) .

Given the overlap in clinical features of fever, aches, leucopenia and thrombocytopenia also observed with dengue fever which is endemic in Singapore (Ministry of Health Singapore, 2020), dengue IgG, IgM and NS1 antigen were evaluated in 35 patients. All 35 patients were found to be negative for dengue co-infection. As J o u r n a l P r e -p r o o f expected, those who were tested for dengue co-infection by the primary physician had clinical features overlapping with dengue fever, in particular, they had lower platelet levels, more pronounced lymphopenia and a higher prevalence of fever than those who were not tested for dengue (Supplementary Table 1) . Coinfection with other vector-bourne diseases such as chikungunya fever and zika virus etc were not tested for as these are rare in Singapore (Ministry of Health Singapore, 2020).

In our entire cohort, 9% developed a pneumonia ( There was no difference in outcome of developing a pneumonia between the different groups of patients with musculoskeletal presentations. Outcome of COVID-19 infection was favourable for these 5 patients with arthralgia and none developed a pneumonia.

To date, studies in COVID-19 have considered the occurrence of myalgias and arthralgias together as a single entity (Cipollaro et al., 2020) . This is the first study that characterizes the various musculoskeletal This made the diagnosis of gout or pseudogout flare less likely. Ours is the first cohort study that specifically describes the arthralgia in COVID-19 patients to be distinct from other musculoskeletal manifestations and not to be attributable to other causes of acute inflammatory arthritis, suggesting this to be a novel clinical manifestation of COVID-19. Dengue co-infection was also unlikely to account for the musculoskeletal symptoms in our cohort.

COVID-related arthralgia appears to be a distinct clinical entity from the generalised bodyache and myalgia which are more commonly described, because it does not conform to the classical prodromal symptoms of a viral infection. Though numbers are small, this group appeared to be older, symptoms were more protracted, and the onset of arthralgia sometimes occurred days before or after the onset of fever and respiratory symptoms. In contrast, myalgia has been reported to coincide with the occurrence of viral illnesses, which is thought to be a result of the acute cytokine response and often resolves after resolution of the fever (Kelvin et al., 2011 , Bian et al., 2014 . In our study, patients with arthralgia also had more severe J o u r n a l P r e -p r o o f pain requiring more analgesia than those with other types of musculoskeletal complaints. In two separate case reports of COVID-19 associated reactive arthritis, one occurred 3 days after and another occurred 20 days after the onset of fever or respiratory symptoms (Liew et al., 2020 , Saricaoglu et al., 2020 . We have also observed a tendency towards an older age in other case reports describing COVID-19 associated arthralgia (Liew et Mechanisms for arthralgia in COVID-19 are currently unknown. Angiotensin-converting enzyme 2, known to be a receptor for entry of SARS-CoV2 into cells, is expressed in multiple extrapulmonary tissue (Hamming et al., 2004) . Direct viral synovial damage is a plausible mechanism for arthralgia, given that ACE-2 was also found to be present in synovial tissue (Mokuda et al., 2020) . However, there has been no studies before demonstrating the presence of SARS-CoV-2 in synovial tissue. In two case report of COVID-19 related reactive arthritis, SARS-CoV-2 was not detected within the synovial fluid (Liew et al., 2020 , Ono et al., 2020 . Immune complex deposition is involved in the pathogenesis of viral arthritis seen in hepatitis B and parvovirus (Wands et al., 1975 , Kerr, 2020 , and has been proposed to contribute to the inflammatory cascade in SARS-CoV-2 infection, although no one has studied its presence in synovial fluid in COVID-19 infection. Transient synovitis or enthesopathy are other possible mechanisms implicated in viral-associated arthralgia (Goupil and Mores, 2016, Whitelaw and Varacallo, 2020). Given the possibility for more than one pathogenic mechanism behind COVID-associated viral arthralgia, the onset has been variably observed to occur days before or after the onset of fever or respiratory symptoms.

As there is a substantial proportion of COVID-19 patients with musculoskeletal symptoms, and a quarter of those with musculoskeletal symptoms had fever without concurrent respiratory symptoms, COVID-19 has to be considered as a differential in equatorial countries such as Singapore, where vector-borne diseases such as dengue or chikungunya are endemic. In the case report by Joob, a Thai lady who presented with fever and arthralgia prior to developing respiratory symptoms was initially misdiagnosed as having dengue fever (Joob and Wiwanitkit, 2020) . This study highlights the importance of recognising musculoskeletal symptoms as a prominent presenting complaint of COVID-19 infection in order to diagnose and institute appropriate isolation measures in a timely fashion.

The results of our study confirm our hypothesis that the presence of musculoskeletal symptoms was not predictive of the risk of developing a pneumonia, consistent with that observed in another study (Lippi et al., 2020) . This is despite a higher prevalence of fever and CRP, a biomarker of inflammation. In addition, none of our patients with arthralgia developed a pneumonia, however the numbers were too small in this group to make a definite conclusion. Larger studies are required to investigate if the presence of arthralgia or musculoskeletal symptoms may indicate a favourable prognosis. Our analysis did demonstrate diabetes to be correlated with an increased risk of developing COVID-19 pneumonia, findings echoed by reports from other centres worldwide (Zhou et al., 2020) .

There are some limitations in our study to be considered. Firstly, all patients were relatively young healthy adult males due to the foreign worker predominance of the outbreak of COVID-19 infections in Singapore.

The study was conducted in patients admitted to the general ward, excluding those who presented with severe manifestations of COVID-19. Hence, these factors may limiting generalizability to other populations.

Secondly, statistical analysis was limited by the low number of patients developing adverse COVID-19 events, reflective of the Singapore's national statistics of one of the world's lowest mortality rate for COVID-19

infection (Worldometers, 2020) . This may have been partly attributable to a milder strain of the virus in J o u r n a l P r e -p r o o f Singapore (Young et al., 2020) , hence our findings may not be applicable to other cohorts worldwide. Thirdly, as most patients had mild musculoskeletal symptoms, further evaluation with synovial fluid analysis or joint ultrasonography was not performed. Lastly, as most musculoskeletal symptoms had resolved in the current admission, there was no follow-up data after discharge, limiting an assessment of recurrence. However, strengths of the study should be acknowledged such as the originality of the clinical question leading to the study. Secondly, all cases of COVID-19 were diagnosed with RT-PCR rather than using a clinical diagnosis. In addition, a younger and relatively healthy cohort is less likely to have comorbidities such as osteoarthritis or gout which can confound the observation of acute arthralgia in COVID-19 hospitalised patients.

The musculoskeletal manifestations of COVID-19 are characterized for the first time in this study.

Reassuringly, we have shown that the presence of musculoskeletal symptoms is not associated with developing a viral pneumonia in COVID-19. COVID-19 associated viral arthralgia in our cohort is a novel clinical entity which does not appear to be typical of a viral prodrome, nor is it typical of a reactive arthropathy, and has distinct characteristics from the other musculoskeletal presentations of COVID-19.

Given the public health risk posed by COVID-19, clinician awareness is important as acute respiratory symptoms may be absent at the time of presentation. Further studies are required to elucidate the underlying mechanism(s) of COVID-19 associated arthralgia, and to ascertain its prognostic implications.

J o u r n a l P r e -p r o o f 

Immune response to SARS-CoV-2 and mechanisms of immunopathological changes in COVID-19

Clinical aspects and cytokine response in adults with seasonal influenza infection

Clinical characteristics of 113 deceased patients with coronavirus disease 2019: retrospective study

Musculoskeletal symptoms in SARS-CoV-2 (COVID-19) patients

Letter from Singapore: The clinical and research response to COVID-19

Infection and musculoskeletal conditions: Viral causes of arthritis

Human Coronavirus Infections in Israel: Epidemiology, Clinical Symptoms and Summer Seasonality of HCoV-HKU1

A Review of Chikungunya Virus-induced Arthralgia: Clinical Manifestations, Therapeutics, and Pathogenesis

Management of post-acute covid-19 in primary care

Clinical Characteristics of Coronavirus Disease 2019 in China

Extrapulmonary manifestations of COVID-19

Increased Complement Receptor-3 levels in monocytes and granulocytes distinguish COVID-19 patients with pneumonia from those with mild symptoms

Tissue distribution of ACE2 protein, the functional receptor for SARS coronavirus. A first step in understanding SARS pathogenesis

Arthralgia as an initial presentation of COVID-19: observation. Rheumatol Int

Inflammatory Cytokine Expression Is Associated with Chikungunya Virus Resolution and Symptom Severity. PLOS Neglected Tropical Diseases

Pathogenesis of human parvovirus B19 in rheumatic disease

COVID-19 patients' clinical characteristics, discharge rate, and fatality rate of meta-analysis

Symptom combinations associated with outcome and therapeutic effects in a cohort of cases with SARS

A Case of Reactive Arthritis Secondary to Coronavirus Disease 2019 Infection

Myalgia may not be associated with severity of coronavirus disease 2019 (COVID-19)

Case series of acute arthritis during COVID-19 admission

Covid-19: What do we know about "long covid

Viral arthritis

Angiotensin-converting Enzyme 2, a SARS-CoV-2 receptor, Is Upregulated by Interleukin 6 through STAT3 Signaling in Synovial Tissues

Middle East respiratory syndrome

Rheumatological Manifestations Associated with Viral Hepatitis B or C

Reactive arthritis after COVID-19 infection

Viral arthritis and COVID-19

The first reactive arthritis case associated with COVID-19

Diagnosis and classification of reactive arthritis

Viral Arthritis

Is routine viral screening useful in patients with recent-onset polyarthritis of a duration of at least 6 weeks? Results from a nationwide longitudinal prospective cohort study

The pathogenesis of arthritis associated with acute hepatitis-B surface antigen-positive hepatitis. Complement activation and characterization of circulating immune complexes

StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2020

Update (Live): 34,481,669 Cases and 1,027,653 Deaths from COVID-19 Virus Pandemic -Worldometer

Effects of a major deletion in the SARS-CoV-2 genome on the severity of infection and the inflammatory response: an observational cohort study

Extrapulmonary complications of COVID-19: A multisystem disease

Comorbidities and the risk of severe or fatal outcomes associated with coronavirus disease 2019: A systematic review and meta-analysis

Ethical approval This study was approved by the Institution Ethics Board of Woodlands Health Campus (DSRB 2020/00765)

This research did not receive any specific grant from funding agencies in the public, commercial, or not-forprofit sectors.

The authors declare that we have no conflict of interest.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.