medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020.

The copyright holder for this preprint

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

1 Multi-organ impairment in low-risk individuals with long COVID

2 Andrea Dennis PhD1, Head of Biomarker Science, Perspectum
3 [email protected]
4 Malgorzata Wamil PhD2, 3, Consultant Cardiologist [email protected]
5 Sandeep Kapur MBBS4, Associate Medical Director [email protected]
6 Johann Alberts MBBCh5, Medical Director [email protected]
7 Andrew D. Badley MD6, Professor of Medicine & Chair, Mayo Clinic COVID Research
8 Taskforce [email protected]
9 Gustav Anton Decker MBBCh6, President, Mayo Clinic International
10 [email protected]
11 Stacey A Rizza6, Professor of Medicine [email protected]
12 Rajarshi Banerjee DPhil*
1,3, Chief Executive and Honorary Consultant Physician
13 [email protected]
14 Amitava Banerjee DPhil*
7,8,9, Associate Professor of Clinical Data Science and Honorary
15 Consultant Cardiologist [email protected]
16 On behalf of the COVERSCAN study investigators (listed at the end of manuscript)
17
1
18 Perspectum, 5520 John Smith Drive, Oxford, OX4 2LL, UK
2
19 Great Western Hospitals NHS Foundation Trust
3
20 Oxford University Hospitals NHS Foundation Trust
4
21 Mayo Clinic Healthcare, 15 Portland Pl, Marylebone, London W1B 1PT
5
22 Alliance Medical Limited, Iceni Centre, Warwick Technology Park, Warwick, CV34 6DA
6
23 Mayo Clinic | 200 First Street SW, Rochester, MN 55905
7
24 Institute of Health Informatics, University College London, 222 Euston Road, London, UK
8
25 University College London Hospitals NHS Trust, 235 Euston Road, London, UK
9
26 Barts Health NHS Trust, The Royal London Hospital, Whitechapel Rd, London, UK
27 *joint senior author
28
29
Corresponding authors: [email protected]; [email protected]
30
31
32
33
34
35
36
37
38
39
40

NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

41 Multi-organ impairment in low-risk individuals with long COVID

42 Abstract

43 Background: Severe acute respiratory syndrome-coronavirus 2 (SARS-CoV-2) infection

44 has disproportionately affected older individuals and those with underlying medical
45 conditions. Research has focused on short-term outcomes in hospital, and single organ
46 involvement. Consequently, impact of long COVID (persistent symptoms three months post-
47 infection) across multiple organs in low-risk individuals is yet to be assessed.

48 Methods: An ongoing prospective, longitudinal, two-centre, observational study was

49 performed in individuals symptomatic after recovery from acute SARS-CoV-2 infection.
50 Symptoms and organ function (heart, lungs, kidneys, liver, pancreas, spleen) were assessed
51 by standardised questionnaires (EQ-5D-5L, Dyspnoea-12), blood investigations and
52 quantitative magnetic resonance imaging, defining single and multi-organ impairment by
53 consensus definitions.

54 Findings: Between April and September 2020, 201 individuals (mean age 44 (SD 11.0)
55 years, 70% female, 87% white, 31% healthcare workers) completed assessments following
56 SARS-CoV-2 infection (median 140, IQR 105-160 days after initial symptoms). The
57 prevalence of pre-existing conditions (obesity: 20%, hypertension: 6%; diabetes: 2%; heart
58 disease: 4%) was low, and only 18% of individuals had been hospitalised with COVID-19.
59 Fatigue (98%), muscle aches (88%), breathlessness (87%), and headaches (83%) were the
60 most frequently reported symptoms. Ongoing cardiorespiratory (92%) and gastrointestinal
61 (73%) symptoms were common, and 42% of individuals had ten or more symptoms.
62 There was evidence of mild organ impairment in heart (32%), lungs (33%), kidneys (12%),
63 liver (10%), pancreas (17%), and spleen (6%). Single (66%) and multi-organ (25%)
64 impairment was observed, and was significantly associated with risk of prior COVID-19
65 hospitalisation (p<0.05).

66 Interpretation: In a young, low-risk population with ongoing symptoms, almost 70% of

67 individuals have impairment in one or more organs four months after initial symptoms of
68 SARS-CoV-2 infection. There are implications not only for burden of long COVID but also
69 public health approaches which have assumed low risk in young people with no
70 comorbidities.

71 Funding: This work was supported by the UK’s National Consortium of Intelligent Medical
72 Imaging through the Industry Strategy Challenge Fund, Innovate UK Grant 104688, and also
73 through the European Union’s Horizon 2020 research and innovation programme under
74 grant agreement No 719445.
75
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

76
77 Introduction
78 Early in the COVID-19 pandemic, research and clinical interest in SARS-CoV-2 (Severe
79 acute respiratory syndrome-coronavirus 2)-induced organ damage was predominantly
80 focused on the respiratory system(1). There have been indirect effects on other organ
81 systems and disease processes, such as cardiovascular diseases and cancers, through
82 changes in health systems or behaviours of patients and health professionals(2-4). In
83 addition, beyond an acute systemic inflammatory response, evidence for direct COVID-19-
84 related effects on multiple organs is accumulating, with potential long-term impacts for
85 individuals as well as health systems(5-8). However, no study to-date has included detailed
86 characterisation of all major organ systems following SARS-CoV-2 infection.
87
88 COVID-19 represents a convergence of an infectious disease, under-treated non-
89 communicable diseases and social determinants of health, as a “syndemic”(9). Pre-existing
90 non-communicable diseases and risk factors are important predictors of poor COVID-19
91 outcomes, whether intensive care admissions or mortality(2). Research has focused on the
92 acute phase of SARS-CoV-2 infection, in hospitalised patients, and on individuals that have
93 died from COVID-19(10-12). It is clear that COVID-19 can have longer multiple symptoms
94 and long-term effects(13), but “long-COVID” is yet to be fully defined(14-15), partly due to
95 lack of understanding of medium- and long-term pathophysiology across organ systems.
96
97 Long COVID in low-risk individuals, who represent up to 80% of the population(2), has public
98 health importance in terms of burden of disease and healthcare utilisation, and therefore has
99 urgent policy relevance across countries. However, in the UK, government policies have
100 emphasised excess risk of mortality in moderate- and high-risk conditions, including
101 “shielding”(2) and commissioning of a risk calculator to identify those at highest risk of
102 COVID-19 severity and mortality(16). As the pandemic progresses, there is growing concern
103 regarding prolonged isolation strategies for people with vulnerable conditions and at highest
104 risk of severe COVID-19 outcomes(17). These approaches have assumed low risk of SARS-
105 CoV-2 infection in younger individuals without underlying conditions, based on their low
106 excess mortality, but without knowledge of the chronic pulmonary and extrapulmonary
107 effects of COVID-19.
108
109 In order to better understand the long-term impact of COVID-19 and ultimately inform
110 preventive measures at health system level, we performed a pragmatic, prospective study in
111 low-risk individuals with symptom assessment, multi-organ magnetic resonance imaging
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

112 (MRI) and blood investigations for inflammatory markers at three months post-COVID-19
113 diagnosis.
114 Methods
115 Patient population and study design
116 In an ongoing, prospective study, 201 participants were enrolled at two UK sites
117 (Perspectum, Oxford and Mayo Clinic Healthcare, London) between April 2020 and August
118 2020 and completed baseline assessment by 14 September 2020 (Figure 1). Participants
119 were eligible for enrolment if they tested positive by the oro/nasopharyngeal throat swab for
120 SARS-CoV-2 by reverse-transcriptase-polymerase-chain reaction (n=62), a positive antibody
121 test (n=63), or had typical symptoms and were determined to have COVID-19 by two
122 independent clinicians (n=73). Exclusion criteria were symptoms of active respiratory viral
123 infection (temperature >37.8°C or three or more episodes of coughing in 24 hours);
124 discharged from hospital in the last 7 days; and contraindications to MRI, including implanted
125 pacemakers, defibrillators, other metallic implanted devices; claustrophobia. The study
126 protocol was approved by a UK ethics committee (20/SC/0185), registered
127 (https://clinicaltrials.gov/ct2/show/NCT04369807) and all patients gave written informed
128 consent.
129
130 To assess the burden of multi-organ involvement after SARS-CoV2 infection
131 Organ function was assessed by patient-reported validated questionnaires, fasting blood
132 investigations (as listed below) and multi-organ MRI. MRI was the chosen imaging modality
133 (as in UK Biobank) because it is: (1) safe, with no radiation exposure, no need for
134 intravenous contrast, minimal contact with the radiographer; (2) quantitative, repeatable and
135 robust, with >95% acquisition and image processing success rate; (3) informative through a
136 repository of digital data which can be shared in the research community for independent
137 analysis and research; (4) rapid and scalable, i.e. a 35-minute scan can phenotype the lung,
138 heart, kidney, liver, pancreas and spleen. At time of MRI, we completed (i) questionnaires for
139 quality of life(EQ-5D-5L(18)), addressing mobility, self-care, usual activity, pain and anxiety,
140 and breathlessness (Dyspnoea-12(19)) and (ii) full blood count, serum biochemistry (sodium,
141 chloride, bicarbonate, urea, creatinine, bilirubin, alkaline phosphatase, aspartate transferase,
142 alanine transferase, lactate dehydrogenase, creatinine kinase, gamma-glutamyl
143 transpeptidase, total protein, albumin, globulin, calcium, magnesium, phosphate, uric acid,
144 fasting triglycerides, cholesterol (total, HDL, LDL), iron, iron-binding capacity (unsaturated
145 and total) and inflammatory markers (erythrocyte sedimentation rate, ESR; high sensitivity-
146 C-Reactive Protein, CRP) (TDL laboratories, London).
147
148 Magnetic Resonance Image Analysis
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

149 Multi-organ MRI data were collected at both study sites (Oxford: MAGNETOM Aera 1.5T,
150 Mayo Healthcare London: MAGNETOM Vida 3T; both from Siemens Healthcare Erlangen,
151 Germany). The COVERSCAN Multiparametic MRI assessment typically required 35mins per
152 patient, including lungs, heart, liver, pancreas, kidneys and spleen by standardised
153 methodology (Supplementary methods).
154
155 Definition of organ impairment
156 MRI-derived measurements from the heart, lungs, kidney, liver, pancreas and spleen were
157 compared with established reference ranges (Table S1) to determine impairment for each
158 organ. An individual organ was classified as impaired if at least one of the metrics calculated
159 for that organ was outside the reference range. Excessive organ fat was not considered as
160 an indicator of impairment on the assumption that this was likely pre-existing and thus
161 treated separately. Organ impairment was defined for each metric according to established
162 cut-offs (Table S1) and was grouped by evidence of: borderline or low ejection fraction and
163 evidence of myocarditis in the heart; reduced pulmonary dynamic measurements in the
164 lungs; elevated cortical T1 in the kidneys; borderline or definite inflammation in the liver and
165 pancreas; and splenomegaly from spleen length.
166
167 Statistical analysis
168 All statistical analyses were performed using R software (version 3.6.1) with a p-value less
169 than 0.05 considered statistically significant. Descriptive statistics were used to summarise
170 baseline participant characteristics. Mean and standard deviation (SD) were used to
171 describe normally distributed-continuous variables, median with interquartile range (IQR) for
172 non-normally distributed, and frequency and percentage for categorical variables.
173
174 Mean difference in quantitative organ metrics between hospitalised versus not hospitalised
175 were compared using the Wilcoxon test, and difference in the counts of the binary outcomes
176 of those with evidence of organ abnormalities compared using Fisher’s test. Multi-organ
177 impairment was defined as impairment to ≥2 organs. Associations between multi-organ
178 impairment and symptoms, comorbidities and pre-existing risk factors were assessed using
179 Spearman’s correlation. Based on the observed differences between hospitalised and non-
180 hospitalised groups, multivariate logistic regression models were used to assess risk factors
181 for COVID-19 hospitalisation.
182
183
184
185
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

186
187
188
189 Results
190 The mean age was 44.0 (SD: 11.0) years. 70% of individuals were female, 87% were white,
191 31% were healthcare workers, 18% had been hospitalised with COVID-19. Assessment
192 (symptoms, blood and MRI) was a median 140 (IQR 105-160) days after initial symptoms.
193 Relevant past medical history included smoking (3%), asthma (18%), obesity (20%),
194 hypertension (6%), diabetes (2%) and heart disease (4%). The hospitalised group were
195 older (p=0.001), had a higher proportion of non-white participants (p=0.038), and were more
196 likely to report ‘inability to walk’ (p=0.01) than non-hospitalised individuals. There were no
197 other significant differences between risk factors or symptoms reported between the groups.
198 The most commonly reported on-going symptoms (regardless of hospitalisation status) were
199 fatigue (98%), muscle ache (88%), shortness of breath (87%) and headache (83%) (Table 1,
200 Figure 2(a)). Ongoing cardiorespiratory (92%) and gastrointestinal (73%) symptoms were
201 common. 99% of individuals had four or more and 42% had ten or more symptoms. 52% of
202 patients reported persistent moderate problems undertaking usual activities (level 3 or
203 greater in the relevant EQ-5D-5L question). 20% reported Dyspnoea-12 ≥15 (equivalent to
204 ~3 on the MRC dyspnoea grade).
205
206 Blood investigations
207 Triglycerides (p=0.002), cholesterol (p=0.021), LDL-cholesterol (p=0.005) and transferrin
208 saturation (p=0.005) were more likely to be abnormal in hospitalised versus non-hospitalised
209 individuals. Mean corpuscular haemoglobin concentration (26%), alanine transferase (14%),
210 lactate dehydrogenase (16%), triglycerides (12%) and cholesterol (42%) were all abnormally
211 high in ≥10% of all individuals (without separation by hospitalisation status). ESR (13%),
212 bicarbonate (13%), uric acid (16%) and high-sensitivity CRP (13%) were abnormally high in
213 in ≥10% of individuals in the hospitalisation group. Bicarbonate (10%), phosphate (13%), uric
214 acid (11%), and transferrin saturation (19%) were abnormally low in ≥10% of individuals
215 (without separation by hospitalisation status) (Table S2).
216
217 Single and multi- organ impairment
218 Impairment was present in the heart in 32% (myocarditis in 11%; systolic dysfunction in
219 23%), lungs in 33%, kidneys in 12%, liver in 10%, pancreas in 17%, and 6% had evidence of
220 splenomegaly (Table 2, Figure 2(b)). 66% of individuals had impairment in one or more
221 organ systems. There was evidence of multi-organ impairment in 25% of individuals, with
222 varying degrees of overlap across multiple organs (Figure 1 and 3). Organ impairment was
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

223 more common in hospitalised versus non-hospitalised individuals. Measures of inflammation

224 in the kidneys and pancreas, and ectopic fat in the pancreas and liver, were also higher in
225 hospitalised individuals (all p <0.05)(Figure 2(b)).
226
227 Association between symptoms, blood investigations and organ impairment
228 Figure 4 shows the percentage of reported symptoms in those with organ impairment (per
229 organ). Multi-organ involvement was associated with more serious symptoms (fatigue,
230 breathlessness etc), but no clear pattern was observed linking symptoms to organ
231 impairment. Regression analysis did not show any association between specific organ
232 impairment and specific symptoms or blood investigations. Increasing age (OR: 1.06 [CI:
233 1.02-1.10], p< 0.01), increased liver volume (OR: 1.18 [CI: 1.06-1.30], p<0.001) and having
234 multi-organ impairment (OR: 2.75 [CI:1.22-6.22], p <0.05), all significantly increased the
235 likelihood of being hospitalized, adjusting for gender and BMI.
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

260
261
262
263 Discussion
264 In the first study to-date evaluating medium-term impairment across multiple organs
265 following SARS-CoV2 infection, we had three major findings. First, in young individuals,
266 largely without risk factors, pre-existing disease or hospitalisation, there was significant
267 symptom burden and evidence of heart, lung, liver and pancreas impairment four months
268 post-COVID-19. Second, symptoms and blood investigations predicted neither organ
269 impairment nor hospitalisation. Third, cardiac (myocarditis and systolic dysfunction) and lung
270 impairment have similar prevalence in low-risk individuals with long COVID.
271
272 The short-term symptoms likely to predict COVID-19(20) persist four months post-infection,
273 particularly fatigue, shortness of breath, myalgia, headache and arthralgia. In this young
274 cohort with low prevalence of comorbidities, the extent of symptom burden and organ
275 impairment is concerning. Models of population COVID-19 impact have been based on age,
276 underlying conditions and mortality, excluding morbidity or potential for multi-organ
277 impairment and chronic diseases(21, 22). Moreover, studies highlighting extrapulmonary
278 COVID-19 manifestations emphasised acute phase of illness(20). Although we describe mild
279 rather than severe organ impairment, the pandemic’s scale and high infection rates in lower
280 risk individuals (by age and underlying conditions), suggest a medium- and longer- term
281 impact of SARS-CoV-2 infection which cannot be ignored in healthcare or policy spheres.
282
283 Although there may be an immunologic basis for variations in progression and severity of
284 SARS-CoV-2 infection in different individuals(24), prediction models to-date have high rates
285 of bias and poor performance(25). We found clustering of cardiorespiratory and
286 gastrointestinal symptoms with evidence of impairment in heart, liver and pancreas
287 respectively, but blood investigations were not associated with particular patterns of organ
288 impairment as determined by COVERSCAN multi-organ assessment. Neither symptoms nor
289 blood investigations were predictive of organ impairment. In acutely unwell patients, the
290 focus has been on recognition of respiratory dysfunction and early provision of ventilatory
291 support, but chronic multi-organ function has not been described systematically. Ongoing
292 studies are considering chronic impact of COVID-19(26) but excluding non-hospitalised, low-
293 risk individuals with and without organ impairment, which we will be investigating further in
294 the longer term. As well as interest in specialist long COVID clinical services(27), there is a
295 role for multi-organ assessment and ongoing evaluation, including low-risk, non-hospitalised
296 individuals, perhaps even in the absence of symptoms.
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

297
298 Acute myocarditis and cardiogenic shock have been described(28), as well as high
299 prevalence of myocarditis in hospitalised COVID-19 patients(29). In American athletes,
300 although recent COVID-19 was associated with myocarditic changes, many non-infected
301 patients also showed these changes(30). We now add that one third of low-risk individuals
302 with long COVID syndrome have cardiac impairment in the form of mild systolic dysfunction
303 or myocarditis three months following SARS-CoV-2 infection. Whilst causality cannot be
304 attributed, cardiac function can be viewed as a risk factor for severe infection and an
305 explanation of persistent symptoms in long COVID. As longitudinal data across organs
306 become available, potential significance of our findings in the liver, kidney and pancreas
307 needs to be explored.

308 Implications for research

309 Our findings at four months post-infection and future findings have three research
310 implications. First, as countries face second pandemic waves, models of the pandemic’s
311 impact must include long COVID, whether quality of life, healthcare utilisation, productivity
312 and economic effects. Second, there is urgent need for further multi-organ assessment,
313 including blood and imaging analysis in the COVID-19 context, as well as linkage with
314 primary and secondary care data, so that long COVID can be properly defined. Third, further
315 longitudinal investigation of clustering of symptoms and organ impairment will inform health
316 services research to plan multidisciplinary care pathways.

317 Implications for clinical practice and public health

318 There are three implications for COVID-19 management. First, as well as highlighting the
319 potential for MRI across organ systems following SARS-CoV-2 infection, our findings signal
320 the need for monitoring and follow-up in at least the medium- and longer-term, especially for
321 extrapulmonary sequelae. Second, as the search for effective COVID-19 vaccines and
322 treatments continues, potential and real long-term multi-organ consequences of SARS-CoV-
323 2 infection in low-risk individuals reinforce the central importance of minimising infection
324 through social distancing, wearing of masks, physical isolation and other population-level
325 measures. Third, both in terms of managing baseline risk, and monitoring and treating
326 complications across organ systems, long COVID requires management across specialities
327 (e.g. cardiology, gastroenterology) and disciplines (e.g. communicable and non-
328 communicable diseases).

329 Strengths and limitations

330 Our study is an ongoing, prospective, longitudinal cohort study with detailed blood and
331 imaging characterisation of organ function, despite limited clinical examination with video
332 consultations in the era of COVID-19. By recruiting ambulatory patients after infection with
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

333 broad inclusion criteria (e.g. SARS-CoV-2 testing by virus RNA, antibody or antigen), we
334 focus on individuals at lower risk of severity and mortality from acute SARS-CoV-2 infection.
335 Our cardiac MRI protocol excluded gadolinium contrast as concerns regarding COVID-19-
336 related renal complications remain. We relied on native T1 mapping to detect and
337 characterise myocardial inflammation, allowing non-invasive tissue characterisation which
338 was previously evaluated as superior to gadolinium MRI for acute myocarditis(31).
339
340 We report baseline findings following SARS-CoV-2 infection. In our pragmatic study design,
341 the diagnosis of COVID-19 was by multiple methods, partly limited by access to laboratory
342 testing during the pandemic. Causality of the relationship between organ impairment and
343 infection cannot be deduced, but may be addressed by longitudinal follow-up of individuals
344 with organ impairment. Our study population was limited by ethnicity despite
345 disproportionate impact of COVID-19 in non-white individuals. Pulse oximetry and spirometry
346 were added later to the protocol and follow up; they were not included from the outset to limit
347 interaction and exposure between trial team and patients. We did not include healthy
348 controls or MRI assessment of brain or muscle function.
349

350 Conclusions
351 Long COVID has a physiological basis, with measurable patient-reported outcomes and
352 organ impairment. Medium- and long-term evaluation and monitoring of multi-organ function
353 beyond symptoms and blood investigations is likely to be required, even in lower risk
354 individuals. Health system responses should emphasise suppression of population infection
355 rates, as well as management of pre- and post-COVID-19 risk factors and chronic diseases.
356
357
358
359
360
361
362
363
364
365
366
367
368
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

369
370
371
372
373 COVERSCAN study investigators
374 Perspectum: Mary Xu, Faezah Sanaei-Nezhad, Andrew Parks, Andrea Borghetto, Matthew
375 D Robson, Petrus Jacobs, John Michael Brady, Carla Cascone, Soubera Rymell, Jacky Law,
376 Virginia Woolgar, Velko Tonev, Claire Herlihy, Rob Suriano, Tom Waddell, Henrike Puchte,
377 Alessandra Borlotti, Arun Jandor, Freddie Greatrex, Robin Jones, Georgina Pitts, Ashleigh
378 West, Marion Maguire, Anu Chandra, Naomi Jayaratne, Dali Wu, Stella Kin, Mike Linsley,
379 Valentina Carapella, Isobel Gordon, George Ralli, John McGonigle, Darryl McClymont,
380 Boyan Ivanov, James Owler, Diogo Cunha, Tatiana Lim, Carlos Duncker, Madison Wagner,
381 Marc Goldfinger, Adriana Roca, Charlotte Erpicum, Matthew David Kelly, Rexford D
382 Newbould, Catherine J Kelly, Andrea Dennis, Sofia Mouchti, Arina Kazimianec, Helena
383 Thomaides-Briers, Rajarshi Banerjee
384 Mayo Clinic: Sandeep Kapur, Louise McLaughlin, Stacey A. Rizza
385 University College London: Amitava Banerjee
386 Great Western Hospitals NHS Foundation Trust: Malgorzata Wamil
387 University of Oxford: Yi-Chun Wang, Tom Waddell

388 Contributorship statement:

389 Study design: AD, SK, RB, JA, SR
390 Patient recruitment: SK, RB, COVERSCAN team
391 Data collection: MW, LM, COVERSCAN team
392 Data analysis: AD, COVERSCAN team, AB
393 Data interpretation: AB, AD, MW, RB
394 Initial manuscript drafting: AB, AD, RB
395 Critical review of early and final versions of manuscript: all authors
396 Specialist input: cardiology (MW, AB); general medicine (RB, ADB, GAD); infectious disease
397 (SAR, ADB); imaging (MR, RB); statistics (AD); epidemiology/public health (AB); primary
398 care (SK); healthcare management (JA).

399 Funding acknowledgements

400 AB is supported by research funding from NIHR, British Medical Association, Astra-Zeneca,
401 UK Research and Innovation, and the Innovative Medicines Initiative-2 (BigData@Heart
402 Consortium, under grant agreement No. 116074, supported by the European Union’s
403 Horizon 2020 research and innovation programme and EFPIA; chaired by DE Grobbee and
404 SD Anker, partnering with 20 academic and industry partners and ESC). This work was
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

405 supported by the UK’s National Consortium of Intelligent Medical Imaging through the
406 Industry Strategy Challenge Fund, Innovate UK Grant 104688, and also through the
407 European Union’s Horizon 2020 research and innovation programme under grant agreement
408 No 719445.
409
410 Research in Context
411
412 Evidence before this study
413 We searched PubMed, medRxiv, bioRxiv, arXiv, and Wellcome Open Research for peer-
414 reviewed articles, preprints, and research reports on long COVID syndrome and medium-
415 and long-term impact of coronavirus disease 2019 (COVID-19), using the search terms
416 “coronavirus”, “COVID-19”, and similar terms, “organ impairment”, “organ function” and
417 “morbidity”, up to September 30, 2020. We found no prior studies of medium- or long-term
418 multi-organ impairment due to COVID-19. Prior studies have considered acute phase of
419 illness and hospitalised patients, focusing on “high-risk” individuals based on age and
420 underlying conditions. Without longer term data including lower risk individuals, full
421 population impact of the pandemic cannot be assessed and health system responses cannot
422 be planned.
423
424 Added value of this study
425 In 201 individuals with low risk for COVID-19 severity and mortality (mean age 44 years,
426 20% obesity, 6% hypertension, 2% diabetes and 4% heart disease, 18% hospitalised), we
427 assessed symptoms, blood investigations and multi-organ magnetic resonance imaging
428 across organ systems, four months following SARS-CoV-2 infection. 99% and 42% had ≥4
429 and ≥10 symptoms respectively. Mild organ impairment was present in at least one organ in
430 66% and in 2 or more organs in 25% of individuals. Multi-organ impairment was associated
431 with hospitalisation.
432
433 Implications of all the available evidence
434 These analyses support strategies to suppress and minimise the infection rate in the
435 population; medium- and long-term follow-up after SARS-CoV-2 infection with detailed
436 evaluation across organ systems; and management of underlying conditions and risk factors
437 before and after infection. For the first time, we provide multi-organ assessment in young,
438 low-risk individuals with long COVID to inform healthcare and policy responses.
439
440
441
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

442
443
444
445
446
447
448 References
449 1. World Health Organization. Clinical management of severe acute respiratory infection
450 (SARI) when COVID-19 disease is suspected. Interim guidance 13 March 2020.
451 https://www.who.int/docs/default-source/coronaviruse/clinical-management-of-novel-
452 cov.pdf
453 2. Banerjee A, Pasea L, Harris S, Gonzalez-Izquierdo A, Torralbo A, Shallcross L,
454 Noursadeghi M, Pillay D, Sebire N, Holmes C, Pagel C, Wong WK, Langenberg C,
455 Williams B, Denaxas S, Hemingway H. Estimating excess 1-year mortality from
456 COVID-19 according to underlying conditions and age in England: a rapid analysis
457 using NHS health records in 3.8 million adults. Lancet May 30;395(10238):1715-1725
458 3. Banerjee A, Chen S, Pasea L, Lai A, Katsoulis M, Denaxas S, Nafilyan V, Williams B,
459 Wong WK, Bakhai A, Khunti K, Pillay D, Noursadeghi M, Wu H, Pareek N, Bromage
460 D, Mcdonagh T, Byrne J, Teo JT, Shah A, Humberstone B, Tang LV, Shah ASV,
461 Rubboli A, Guo Y, Hu Y, Sudlow CLM, Lip GYH, Hemingway H. Excess deaths in
462 people with cardiovascular diseases during the COVID-19 pandemic. Medrxiv.
463 Preprint. 2020. Online 11/6/2020.
464 https://www.medrxiv.org/content/10.1101/2020.06.10.20127175v1
465 4. Lai AG, Pasea L, Banerjee A, Denaxas S, Katsoulis M, Chang WH, Williams B, Pillay
466 D, Noursadeghi M, Swanton C, Linch D, Hughes D, Forster MD, Johnson P, Turnbull
467 C, DATA-CAN, Cooper M, Jones M, Pritchard-Jones K, Sullivan R, Lawler M, Hall G,
468 Davie C, Hemingway H. Estimating excess mortality in people with cancer and
469 multimorbidity in the COVID-19 emergency. BMJ Open. 2020. In press.
470 5. Pavon AG, Meier D, Samim D, Rotzinger DC, Fournier S, Marquis P, Monney P,
471 Muller O, Schwitter J. First Documentation of Persistent SARS-Cov-2 Infection
472 Presenting With Late Acute Severe Myocarditis. Can J Cardiol. 2020
473 Aug;36(8):1326.e5-1326.e7.
474 6. Alqahtani SA, Schattenberg JM. Liver injury in COVID-19: The current evidence.
475 United European Gastroenterol J. 2020 Jun;8(5):509-519.
476 7. Farouk SS, Fiaccadori E, Cravedi P, Campbell KN. COVID-19 and the kidney: what
477 we think we know so far and what we don't. J Nephrol. 2020 Jul 20:1-6.
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

478 8. Somasundaram NP, Ranathunga I, Ratnasamy V, Wijewickrama PSA, Dissanayake

479 HA, Yogendranathan N, Gamage KKK, de Silva NL, Sumanatilleke M, Katulanda P,
480 Grossman AB. The Impact of SARS-Cov-2 Virus Infection on the Endocrine System.
481 J Endocr Soc. 2020 Jul 2;4(8):bvaa082.
482 9. Horton R. Offline: COVID-19 is not a pandemic. Lancet 2020. 396; 874.
483 10. Shovlin CL, Vizcaychipi MP. Implications for COVID-19 triage from the ICNARC
484 report of 2204 COVID-19 cases managed in UK adult intensive care units. Emerg
485 Med J. 2020 Jun;37(6):332-333.
486 11. Docherty AB, Harrison EM, Green CA, Hardwick HE, Pius R, Norman L, Holden KA,
487 Read JM, Dondelinger F, Carson G, Merson L, Lee J, Plotkin D, Sigfrid L, Halpin S,
488 Jackson C, Gamble C, Horby PW, Nguyen-Van-Tam JS, Ho A, Russell CD, Dunning
489 J, Openshaw PJ, Baillie JK, Semple MG; ISARIC4C investigators. Features of
490 20 133 UK patients in hospital with covid-19 using the ISARIC WHO Clinical
491 Characterisation Protocol: prospective observational cohort study. BMJ. 2020 May
492 22;369:m1985. doi: 10.1136/bmj.m1985.
493 12. Williamson EJ, Walker AJ, Bhaskaran K, Bacon S, Bates C, Morton CE, Curtis HJ,
494 Mehrkar A, Evans D, Inglesby P, Cockburn J, McDonald HI, MacKenna B, Tomlinson
495 L, Douglas IJ, Rentsch CT, Mathur R, Wong AYS, Grieve R, Harrison D, Forbes H,
496 Schultze A, Croker R, Parry J, Hester F, Harper S, Perera R, Evans SJW, Smeeth L,
497 Goldacre B. Factors associated with COVID-19-related death using OpenSAFELY.
498 Nature. 2020 Jul 8. doi: 10.1038/s41586-020-2521-4.
499 13. World Health Organization. What we know about Long-term effects of COVID-19.9
500 September 2020. https://www.who.int/docs/default-source/coronaviruse/risk-comms-
501 updates/update-36-long-term-symptoms.pdf?sfvrsn=5d3789a6_2
502 14. Nabavi N. Long covid: How to define it and how to manage it. BMJ. 2020 Sep
503 7;370:m3489.
504 15. Greenhalgh T, Knight M, A'Court C, Buxton M, Husain L. Management of post-acute
505 covid-19 in primary care. BMJ. 2020 Aug 11;370:m3026.
506 16. National Institute for Health Research. New risk prediction model could help improve
507 guidance for people shielding from COVID-19. 23 June 2020.
508 https://www.nihr.ac.uk/news/new-risk-prediction-model-could-help-improve-guidance-
509 for-people-shielding-from-covid-19/25096
510 17. Wise J. Covid-19: Experts divide into two camps of action—shielding versus blanket
511 policies. BMJ 2020;370:m3702.https://www.bmj.com/content/370/bmj.m3702
512 18. Janssen MF, Pickard AS, Golicki D, Gudex C, Niewada M, Scalone L, Swinburn P,
513 Busschbach J. Measurement properties of the EQ-5D-5L compared to the EQ-5D-3L
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

514 across eight patient groups: a multi-country study. Qual Life Res 2013
515 Sep;22(7):1717-1727
516 19. Yorke J, Moosavi SH, Shuldham C, Jones PW. Quantification of dyspnoea using
517 descriptors: development and initial testing of the Dyspnoea-12. Thorax. 2010
518 Jan;65(1):21-6. doi: 10.1136/thx.2009.118521. Epub 2009 Dec 8.
519 20. Menni C, Valdes AM, Freidin MB, Sudre CH, Nguyen LH, Drew DA, Ganesh S,
520 Varsavsky T, Cardoso MJ, El-Sayed Moustafa JS, Visconti A, Hysi P, Bowyer RCE,
521 Mangino M, Falchi M, Wolf J, Ourselin S, Chan AT, Steves CJ, Spector TD. Real-
522 time tracking of self-reported symptoms to predict potential COVID-19. Nat Med.
523 2020 Jul;26(7):1037-1040.
524 21. Gupta A, Madhavan MV, Sehgal K, Nair N, Mahajan S, Sehrawat TS, Bikdeli B,
525 Ahluwalia N, Ausiello JC, Wan EY, Freedberg DE, Kirtane AJ, Parikh SA, Maurer
526 MS, Nordvig AS, Accili D, Bathon JM, Mohan S, Bauer KA, Leon MB, Krumholz HM,
527 Uriel N, Mehra MR, Elkind MSV, Stone GW, Schwartz A, Ho DD, Bilezikian JP,
528 Landry DW. Extrapulmonary manifestations of COVID-19. Nat Med. 2020
529 Jul;26(7):1017-1032. doi: 10.1038/s41591-020-0968-3.
530 22. Palmer K, Monaco A, Kivipelto M, Onder G, Maggi S, Michel JP, Prieto R, Sykara G,
531 Donde S. The potential long-term impact of the COVID-19 outbreak on patients with
532 non-communicable diseases in Europe: consequences for healthy ageing. Aging Clin
533 Exp Res. 2020 Jul;32(7):1189-1194.
534 23. Wyper GMA, Assunção R, Cuschieri S, Devleeschauwer B, Fletcher E, Haagsma JA,
535 Hilderink HBM, Idavain J, Lesnik T, Von der Lippe E, Majdan M, Milicevic MS, Pallari
536 E, Peñalvo JL, Pires SM, Plaß D, Santos JV, Stockton DL, Thomsen ST, Grant I.
537 Population vulnerability to COVID-19 in Europe: a burden of disease analysis. Arch
538 Public Health. 2020 May 29;78:47.
539 24. Mathew D, Giles JR, Baxter AE, Oldridge DA, Greenplate AR, Wu JE, Alanio C, Kuri-
540 Cervantes L, Pampena MB, D'Andrea K, Manne S, Chen Z, Huang YJ, Reilly JP,
541 Weisman AR, Ittner CAG, Kuthuru O, Dougherty J, Nzingha K, Han N, Kim J,
542 Pattekar A, Goodwin EC, Anderson EM, Weirick ME, Gouma S, Arevalo CP, Bolton
543 MJ, Chen F, Lacey SF, Ramage H, Cherry S, Hensley SE, Apostolidis SA, Huang
544 AC, Vella LA; UPenn COVID Processing Unit, Betts MR, Meyer NJ, Wherry EJ. Deep
545 immune profiling of COVID-19 patients reveals distinct immunotypes with therapeutic
546 implications. Science. 2020 Sep 4;369(6508):eabc8511.
547 25. Wynants L, Van Calster B, Collins GS, Riley RD, Heinze G, Schuit E, Bonten MMJ,
548 Damen JAA, Debray TPA, De Vos M, Dhiman P, Haller MC, Harhay MO, Henckaerts
549 L, Kreuzberger N, Lohman A, Luijken K, Ma J, Andaur CL, Reitsma JB, Sergeant JC,
550 Shi C, Skoetz N, Smits LJM, Snell KIE, Sperrin M, Spijker R, Steyerberg EW, Takada
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

551 T, van Kuijk SMJ, van Royen FS, Wallisch C, Hooft L, Moons KGM, van Smeden M.
552 Prediction models for diagnosis and prognosis of covid-19 infection: systematic
553 review and critical appraisal. BMJ. 2020 Apr 7;369:m1328. doi: 10.1136/bmj.m1328.
554 26. PHOSP-COVID: Post-HOSPitalisation COVID-19 study. https://www.phosp.org/
555 27. NHS to offer ‘long covid’ sufferers help at specialist centres. 7 October 2020
556 https://www.england.nhs.uk/2020/10/nhs-to-offer-long-covid-help/
557 28. Chau VQ, Giustino G, Mahmood K, Oliveros E, Neibart E, Oloomi M, Moss N, Mitter
558 SS, Contreras JP, Croft L, Serrao G, Parikh AG, Lala A, Trivieri MG, LaRocca G,
559 Anyanwu A, Pinney SP, Mancini DM. Cardiogenic Shock and Hyperinflammatory
560 Syndrome in Young Males with COVID-19. Circ Heart Fail. 2020 Aug 26. doi:
561 10.1161/CIRCHEARTFAILURE.120.007485. Online ahead of print.
562 29. Puntmann VO, Carerj ML, Wieters I, et al. Outcomes of Cardiovascular Magnetic
563 Resonance Imaging in Patients Recently Recovered From Coronavirus Disease
564 2019 (COVID-19). JAMA Cardiol. Published online July 27, 2020.
565 doi:10.1001/jamacardio.2020.3557.
566 30. Rajpal S, Tong MS, Borchers J, et al. Cardiovascular Magnetic Resonance Findings
567 in Competitive Athletes Recovering From COVID-19 Infection. JAMA Cardiol.
568 Published online September 11, 2020. doi:10.1001/jamacardio.2020.4916.
569 31. Ferreira VM, Piechnik SK, Dall'Armellina E, Karamitsos TD, Francis JM, Ntusi N, et
570 al. T1 mapping for the diagnosis of acute myocarditis using CMR: comparison to T2-
571 weighted and late gadolinium enhanced imaging. JACC Cardiovasc Imaging.
572 2013;6(10):1048-58)
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

588
589
590
591
592
593
594
595
596 Figures and Tables
597
598 Table 1: Baseline demographics and symptoms in 201 low-risk individuals with long-COVID.

599 Table 2: Evidence of organ impairment in 201 low-risk individuals with long-COVID.

600
601 Figure 1: Natural history of long COVID, the COVERSCAN study in low-risk individuals
602 (n=201) and policy recommendations.

603 Figure 2: Proportion of low-risk individuals with long-COVID by hospitalisation (n=201) for (a)
604 symptoms; and (b) evidence of organ impairment.

605 Figure 3: Multi-organ impairment in low-risk individuals with long COVID by gender and
606 hospitalisation (n=201).

607
608 Figure 4: Clustering of reported symptoms and organ impairment for individuals with long-
609 COVID (n=201).

610
611
612
613
614
615
616
617
618
619
620
621
622
623
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

624
625
626
627
628
629 Table 1: Baseline demographics and symptoms in 201 low-risk individuals with long-COVID.*

All Not hospitalised Hospitalised p

630
(n=201) (n=164) (n=37) 631
*
N (%) N(%) N(%) 632 Con
Patient 633 tinu
characteristics 634 ous
Age (yrs, mean; sd) 44(11.0) 43(10.9) 50(10.0) 635
0.001 data
Female (No, %) 140(69.7) 117(71.3) 23(62.2) 636
0.218 pres
-2
BMI (kg.m , median; 637 ente
IQR) 25.7(22.7,28.1) 25.3(22.6,27.7) 27.2(23.1,31.0) 638
0.162 d as
Ethnicity 639 mea
White 174(86.6) 146(89.0) 28 (75.7) 0.038
640 ns
Mixed 3(1.5) 3(1.8) 0 (0) 641 (SD)
South Asian 8(4.0) 5(3.0) 3 (8.1)
Black 5(2.5) 3(1.8) 2 (5.4)
642 for
643 nor
Comorbidities and 644 mall
risks
645 y
Smoking 646 distri
Never 132 (65.7) 108 (65.9) 24 (64.9)
Current 6 (3.0) 6 (3.7) 0 (0.0)
647
0.945
bute
Ex 63 (31.3) 50 (30.5) 13 (35.1)
648 d
Health care worker 62 (30.8) 49 (29.9) 13 (35.1) 649
0.557 data
Asthma 36 (17.9) 33(20.1) 3 (8.1) 650
0.099 and
BMI 651 med
≥25 kg/m2 112 (56.3) 87 (53.7) 25 (67.6) 652
0.144 ian
≥30 kg/m2 40 (20.1) 28 (17.3) 12 (32.4) 653
0.103 (IQR
Hypertension 12 (6.0) 10 (6.1) 2 (5.4) 6541 ) for
Diabetes 4 (2.0) 4 (2.4) 0 (0.0) 6551 non-
Previous heart 656 nor
disease 8 (4.0) 7 (4.3) 1 (2.7) 6571 mall
Symptoms 658 y
Fatigue 197 (98.0) 160 (97.6) 37 (100.0) 6591 distri
Muscle ache 176 (87.6) 145 (88.4) 31 (83.8) 0.418
660 bute
Shortness of breath 175 (87.1) 140 (85.4) 35 (94.6) 0.177
661 d
Headache 166 (82.6) 139 (84.8) 27 (73.0) 0.097
662 data
Joint pain 157 (78.1) 128 (78.0) 29 (78.4) 6631 and
Fever 151 (75.1) 127 (77.4) 24 (64.9) 0.14
Chest pain 147 (73.1) 116 (70.7) 31 (83.8)
664
0.15
cate
Cough 148 (73.6) 119 (72.6) 29 (78.4)
665
0.54
gori
Sore throat 143 (71.1) 120 (73.2) 23 (62.2) 666
0.228 cal
Diarrhoea 119 (59.2) 92 (56.1) 27 (73.0) 667
0.066 as
Abnormal pain 108 (53.7) 91 (55.5) 17 (45.9) 668
0.362 cou
Wheezing 97 (48.3) 74 (45.1) 23 (62.2) 669
0.07 nt
Inability to walk 81 (40.3) 59 (36.0) 22 (59.5) 670
0.01 (%).
Runny nose 68 (33.8) 55 (33.5) 13 (35.1) 671
0.85 Com
Time interval 672 pari
Initial symptoms-to- 673 sons
assessment (days: (n=1 missing) (n=1 missing 674 betw
median, [IQR]) 140 (105, 160) 140 (106, 162) 138 (97, 150) 0.002
675 een
COVID-19 positive- 676 pati
to-assessment (days: (n=3 missing) (n=3 missing) 677 ents
median, [IQR]) 70 (42, 112) 67 (39, 109) 105 (59, 126) 0.105
678 man
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

679 aged at home vs hospitalised were conducted using Wilcoxon Rank sum test for continuous data and
680 Fisher exact test for categorical data.
681
682
683
684
685 Table 2: Evidence of organ impairment in 201 low-risk individuals with long-COVID.

Measurement All Not Hospitalised p

(n=201) hospitalised (n=37)
N(%) (n=164) N(%)
N(%)
HEART
Left ventricular ejection fraction
(%)
• Normal (>55%) 155 (77.1) 129 (78.7) 26 (70.3)
• Borderline impairment (50-55%) 38 (18.9) 31 (18.9) 7 (18.9) 0.079
• Definite impairment (<50%) 8 (4.0) 4 (2.4) 4 (10.8)
Left ventricular end diastolic
volume (ml)
• >214ml in M; >178ml in W 27 (13.4) 18 (11.0) 9 (24.3) 0.057
Evidence of myocarditis
• ≥ 3 segments with high T1
(≥1264ms at 3T; ≥1015ms at
1.5T) 22 (10.9) 18 (11.0) 4 (10.8) 1
LUNGS
Deep Breathing Fractional area
change (n= 11 missing) (n= 8 missing) (n= 3 missing)
• < 39% 63 (33.2) 47 (30.1) 16 (47.1) 0.071
KIDNEYS
Kidney cortex T1 (n= 12 missing) (n= 8 missing) (n= 4 missing)
• Normal (<1610 ms at 3T; 0.046
<1191ms at 1.5T) 175 (88.4) 146 (90.7) 29 (78.4)
• Definite impairment (≥1610ms at
3T; ≥1191ms at 1.5T ) 23 (11.6) 15 (9.3) 8 (21.6)
PANCREAS
Pancreatic inflammation (T1 in ms)
• Normal (<800ms) 157 (83.1) 136 (87.2) 21 (63.6)
• Borderline (800-865ms) 20 (10.6) 11 (7.1) 9 (27.3) 0.003
• Significant (>865ms) 12 (6.3) 9 (5.8) 3 (9.1)
Pancreatic fat (n= 6 missing) (n= 4 missing) (n= 2 missing)
• Normal (<5%) 126 (64.6) 111 (69.4) 15 (42.9)
0.005
• Borderline (5-10%) 44 (22.6) 33 (20.6) 11 (31.4)
• Significant(>10%) 25 (12.8) 16 (10.0) 9 (25.7)
LIVER
Liver Inflammation (cT1 in ms) (n= 1 missing) (n= 1 missing)
• Normal (<800ms) 181 (90.5) 150 (92.0) 31 (83.8)
• Borderline (800-825ms) 5 (2.5) 5 (3.1) 0 (0.0) 0.040
• Significant (>825ms) 14 (7.0) 8 (4.9) 6 (16.2)
Liver fat
• Normal (<5%) 162 (80.6) 138 (84.1) 24 (64.9)
• Borderline (5-10%) 18 (9.0) 12 (7.3) 6 (16.2)
0.025
• Definite (>10%) 21 (10.4) 14 (8.5) 7 (18.9)
SPLEEN
Splenic length (mm) (n= 10 missing) (n= 10 missing)
• Normal (Table S1) 179(9.4) 144(9.5) 35 (9.5)
• Borderline (Table S1) 12 (6.3) 10 (6.5) 2 (5.4) 1
686
*
687 Data are presented as count (%). Comparisons between patients managed at home vs hospitalised
688 were conducted using Fisher exact test.
689
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .
Figure 1: Natural history of long COVID, the COVERSCAN study in low-risk individuals (n=201) and policy recommendations.
690

691
692
693
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

694 Figure 2: Proportion of low-risk individuals with long-COVID by hospitalisation (n=201) for (a)
695 symptoms; and (b) evidence of organ impairment.

696
697 (a)

698
699 (b)

700
701
702
703
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

704 Figure 3 Multi-organ impairment in low-risk individuals with long COVID by gender and
705 hospitalisation (n=201).

706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
medRxiv preprint doi: https://doi.org/10.1101/2020.10.14.20212555.this version posted October 16, 2020. The copyright holder for this preprint
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
It is made available under a CC-BY-NC-ND 4.0 International license .

727 Figure 4: Clustering of reported symptoms and organ impairment for individuals with long-
728 COVID (n=201).

729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747