Stroke

BRIEF REPORT

Acute Stroke in Times of the COVID-19

Pandemic
A Multicenter Study
Carolin Hoyer , MD; Anne Ebert, PhD; Hagen B. Huttner, MD; Volker Puetz, MD; Bernd Kallmünzer, MD;
Kristian Barlinn, MD; Christian Haverkamp, MD; Andreas Harloff, MD; Jochen Brich, MD; Michael Platten, MD; Kristina Szabo, MD

BACKGROUND AND PURPOSE: This study aims to assess the number of patients with acute ischemic cerebrovascular events
seeking in-patient medical emergency care since the implementation of social distancing measures in the coronavirus
disease 2019 (COVID-19) pandemic.

METHODS: In this retrospective multicenter study, data on the number of hospital admissions due to acute ischemic stroke
or transient ischemic attack and numbers of reperfusion therapies performed in weeks 1 to 15 of 2020 and 2019 were
collected in 4 German academic stroke centers. Poisson regression was used to test for a change in admission rates before
and after the implementation of extensive social distancing measures in week 12 of 2020. The analysis of anonymized
regional mobility data allowed for correlations between changes in public mobility as measured by the number and length of
trips taken and hospital admission for stroke/transient ischemic attack.

RESULTS: Only little variation of admission rates was observed before and after week 11 in 2019 and between the weeks 1
and 11 of 2019 and 2020. However, reflecting the impact of the COVID-19 pandemic, a significant decrease in the number
of admissions for transient ischemic attack was observed (−85%, −46%, −42%) in 3 of 4 centers, while in 2 of 4 centers,
Downloaded from http://ahajournals.org by on October 3, 2021

stroke admission rates decreased significantly by 40% and 46% after week 12 in 2020. A relevant effect on reperfusion
therapies was found for 1 center only (thrombolysis, −60%; thrombectomy, −61%). Positive correlations between number of
ischemic events and mobility measures in the corresponding cities were identified for 3 of 4 centers.

CONCLUSIONS: These data demonstrate and quantify decreasing hospital admissions due to ischemic cerebrovascular events
and suggest that this may be a consequence of social distancing measures, in particular because hospital resources for acute
stroke care were not limited during this period. Hence, raising public awareness is necessary to avoid serious healthcare and
economic consequences of undiagnosed and untreated strokes and transient ischemic attacks.

Key Words:  brain ischemia ◼ pandemics ◼ reperfusion ◼ stroke ◼ thrombectomy

T
he ongoing coronavirus disease 2019 (COVID- stroke-related deaths, and disability-adjusted life-years
19) pandemic poses unprecedented challenges to are globally increasing. In high-income countries, tremen-
healthcare systems worldwide. Infections and fatali- dous efforts in improving prevention, acute treatment, and
ties are continuously rising, necessitating restructuring neurorehabilitation have led to a substantial decrease
of health services and the reallocation of resources in in the burden of stroke over the past 30 years.1 During
many places. the recent spread of COVID-19, neurologists worldwide
Stroke is the second leading cause of death world- have reported the impression of an apparent reduction of
wide, and absolute numbers of incident strokes, survivors, numbers of patients with acute stroke seeking medical

 
Correspondence to: Carolin Hoyer, MD, Department of Neurology, Universitätsmedizin Mannheim, Theodor-Kutzer-Ufer 1-3, 68167 Mannheim, Germany. Email carolin.
[email protected]
This manuscript was sent to Marc Fisher, Senior Consulting Editor, for review by expert referees, editorial decision, and final disposition.
For Sources of Funding and Disclosures, see page 2227.
© 2020 American Heart Association, Inc.
Stroke is available at www.ahajournals.org/journal/str

2224   July 2020 Stroke. 2020;51:2224–2227. DOI: 10.1161/STROKEAHA.120.030395

 Hoyer et al Stroke and COVID-19

emergency care.2 Reasons for this phenomenon are not the impact of the COVID-19 pandemic). This effect is expressed
understood although some suggest that patients’ fear of as an incidence rate ratio along with its 2-sided 95% CI:
in-hospital infection causes avoidance behavior. incidence rate 2020

Brief Report
incidence rate ratio =
The objective of this study was to assess and quantify incidence rate 2019
the dynamics of admission for stroke and transient isch-
where
emic attack (TIA) and rates of reperfusion therapies dur-
ing the COVID-19 pandemic and to estimate a potential number of events in late epoch 2020
incidence rate 2020 =
impact of the partial lockdown in Germany. number of events in early epoch2020

and where
number of events in late epoch 2019
METHODS incidence rate 2019 =
number of events in early epoch2019
This retrospective cohort study was performed at 4 German
comprehensive stroke centers (University Hospital Dresden, For the covariance matrix, a robust estimator was chosen to
University Erlangen-Nuremberg, Medical Center University of account for possible overdispersion.
Freiburg, University Medical Center Mannheim; A–D, not in Rank correlations (Spearman rho) were calculated to ana-
alphabetic order), collectively treating about 6000 stroke and lyze correlations between mobility and admission for isch-
TIA patients per year. The study was approved by the respective emic cerebrovascular events. P<0.05 indicates statistical
local ethics committees. Written informed consent was waived significance.
due to the retrospective character of the investigation. Statistical analysis was performed using IBM SPSS
Numbers of patients admitted with final diagnoses of isch- Statistics, version 25.
emic stroke or TIA between January 1, 2019, April 14, 2019
(weeks 1–15), and between January 1, 2020, and April 12,
2020 (weeks 1–15), and information regarding reperfusion RESULTS
therapies (intravenous thrombolysis and mechanical throm-
bectomy) were gathered from main discharge diagnosis docu-
Only little variation of admission rates was observed
mentation according to the German diagnosis-related group before and after week 11 in 2019 and between weeks
system. Treatment was defined by Operation and Procedure 1 and 11 in 2019 and in 2020. Between early (weeks
Classification System codes for intravenous thrombolysis and 1—11) and late (weeks 12—15) investigational peri-
thrombectomy. ods in 2019 and 2020, in all but center B, a signifi-
Week 12, beginning March 16, 2020, was defined as the cant decrease in the number of admissions for TIA was
Downloaded from http://ahajournals.org by on October 3, 2021

week when the COVID-19 pandemic began to impact public observed (center A: -85%; center C: -43%; center D:
life in Germany. Extended measures for social distancing, for -46%). Moreover, in centers A and D, stroke admis-
example, through closure of schools/day care and partial lock- sion rates decreased significantly by 38% and 46%,
down in all German federal states, were implemented beginning as reflected by incidence rate ratios of 0.62 and 0.54,
March 16, 2020, and were strongly advocated by the German
respectively (Table). A relevant effect of the year-by-
Chancellor in her television address on March 18, 2020.
epoch interaction on reperfusion therapies was found
Anonymized mobility data were obtained from Teralytics,
Zurich, Switzerland, and include the number of trips and for center A with a significant drop in the thrombolysis
kilometers traveled in Dresden, Erlangen, Freiburg, and rate by 60% and in the thrombectomy rate by 61%. For
Mannheim. Mobility data are generated through the registra- details, see the Table. Differences of cumulative admis-
tion of mobile phones with cell towers. Movements are reg- sions for acute stroke and TIA in the 4 study centers
istered when an individual switches cell tower areas. They between weeks 10 and 15 in 2020 and weeks 10 and
are attributed to special aggregation areas. Mobility flows are 15 in 2019, as well as changes in mobility throughout
generated from the aggregation of all movements in a certain the observation period in 2020, are shown in the Figure.
time frame. Information regarding the local case numbers of From week 12 onward, a pronounced decrease in
COVID-19 infections was obtained from the Robert Koch public mobility was observed in all 4 cities. Positive corre-
Institute.3
lations between the number of admissions due to stroke/
Anonymized data will be shared by request from any quali-
TIA and measures of mobility in the corresponding cities
fied investigator.
were found for 3 of the 4 centers (center A—number of
trips: rs=0.54, P=0.045; kilometers traveled: rs=0.54,
Statistical Analysis P=0.047; center C—number of trips: rs=0.57, P=0.032;
Calendar weeks 1 to 11 and 12 to 15 were categorized into a kilometers traveled: rs=0.54, P=0.045; center D—kilome-
variable epoch for the years 2019 and 2020. The dependent ters traveled: rs=0.61, P=0.020).
variables were count of admitted strokes, TIAs, intravenous
thrombolysis, and thrombectomy. Poisson regression was used
to test whether the rate of events (admissions for stroke/TIA,
intravenous thrombolysis, thrombectomy) changed as a func-
DISCUSSION
tion of year (2019–2020), epoch (weeks 1–11, early; weeks This study identified an impact of the COVID-19 pan-
12–15, late), and the interaction of year and epoch (reflecting demic on admissions for acute stroke and TIA and

Stroke. 2020;51:2224–2227. DOI: 10.1161/STROKEAHA.120.030395 July 2020   2225

 Hoyer et al Stroke and COVID-19

Table.  IRR of Stroke Admissions and Reperfusion Therapies disease-associated predisposition for thromboembolic
Between Early (Weeks 1–11) and Late (Weeks 12–15) events.4
Investigational Periods in 2019 and 2020 in Centers A to D
The 4 centers were affected differently: 3 of these
Brief Report

Center hospitals are in federal states where the pandemic hit

Variable A B C D early and most severely (81.38, 89.76, 237.90 infec-
IRR stroke 0.62 1.16 1.05 0.54 tions/100 000 inhabitants during weeks 12–15, as
  95% CI 0.38–0.99 0.83–1.63 0.74–1.49 0.35–0.83
reported by the Robert Koch Institute). In one center,
0.046* 0.377 0.779 0.005*
no significant impact of the COVID-19 pandemic was
  P value
noted, which may be partly attributable to its location
IRR TIA 0.15 0.91 0.58 0.54
in a state with lower numbers of COVID-19 cases
  95% CI 0.04–0.51 0.47–1.75 0.34–0.99 0.30–0.95 (56.39 infections/100 000 inhabitants during weeks
  P value 0.003* 0.775 0.047* 0.033* 12–15). It bears noting that in none of these centers
IRR IVT 0.40 0.90 1.00 0.78 were measures in place that changed the structure of
  95% CI 0.19–0.86 0.34–2.43 0.48–2.1 0.29–2.13 pre- or early in-hospital stroke care during the pan-
  P value 0.019* 0.838 0.995 0.629 demic. Moreover, at no point since the beginning of the
pandemic has there been a shortage of capacities for
IRR MTE 0.39 1.88 1.21 0.71
admission or patient care.
  95% CI 0.17–0.87 0.87–4.04 0.48–2.99 0.31–1.67
While local characteristics may influence admission
  P value 0.022* 0.108 0.685 0.436
and reperfusion therapy rates, for example, through
IRR indicates incidence rate ratio; IVT, intravenous thrombolysis; MTE, existing stroke networks or the size of a stroke center’s
mechanical thrombectomy; and TIA, transient ischemic attack. catchment area, the decision to seek medical help in
*P<0.05 indicates statistical significance.
the first place, however, is independent of these factors.
Alongside fear of infection, the confinement of patients
verifies observations recently made by neurologists in to their home with family members as primary part-
different countries.2 There are no reasons to assume ners of social interaction may result in communication
that the incidence of stroke is decreasing, if anything, sequences with delayed symptom disclosure and valida-
it may rise in the context of the pandemic, given the tion of watch-and-wait strategies.5
Downloaded from http://ahajournals.org by on October 3, 2021

Figure. Difference (in percentage) of cumulative admissions for acute stroke (green) and transient ischemic attacks (TIA; blue)
in 4 German comprehensive stroke centers between weeks 10 to 15 in 2020 and weeks 10 to 15 in 2019 (positive y axis).
Difference (in percentage) of number of trips (red) and kilometers traveled (orange) in the corresponding cities between weeks 2 to 11 and
weeks 12 to 15 in 2020 (negative y axis). WHO indicates World Health Organization.

2226   July 2020 Stroke. 2020;51:2224–2227. DOI: 10.1161/STROKEAHA.120.030395

 Hoyer et al Stroke and COVID-19

It is noteworthy that reperfusion therapy rates, reflecting Erlangen, Germany (H.B.H., B.K.). Department of Neurology, Dresden Neurovas-
cular Center, Carl Gustav Carus University Hospital, Technische Universität Dres-
severe strokes, remained largely unaffected. This indicates den, Germany (V.P., K.B.). Department of Neurology and Clinical Neuroscience
that both patients and prehospital medical staff correctly

Brief Report
(A.H., J.B.) and Institute of Digitalization in Medicine (C. Haverkamp), Faculty of
recognize the need for urgent assessment and treatment Medicine, University of Freiburg, Germany.
despite the threat of coronavirus infection. However, with
Acknowledgments
the decrease in the number of admissions most noticeable We thank Professor Matthias Gondan, Department of Psychology, University of
for TIA, the fear of infection and adhering to advice from Copenhagen, Denmark, for statistical advice.
health organizations and politicians likely outweighs the
concern over transient or mild neurological dysfunction. Sources of Funding
Dr Hoyer receives a grant for postdoctoral lecture qualification within the Olympia
Decreased public mobility may reflect the effective- Morata Program of Heidelberg University.
ness of stay-at-home orders as one constituent mea-
sure of social distancing. As indicated by the positive Disclosures
correlation of lower admission rates for stroke/TIA with Dr Huttner reports research grants by Novartis, Medtronic, UCB Pharma, and
Portola Pharmaceuticals. Dr Huttner reports personal fees from Bayer AG, Boeh-
decreased public mobility during the pandemic, such ringer Ingelheim, Daiichi Sankyo, Medtronic, CLS Behring, UCB Pharma, and
measures may influence patients’ willingness to seek Portola Pharmaceuticals. Dr Kallmünzer reports speaker honoraria from Daiichi
medical help for acute symptoms. On a related note, Sankyo, Novartis, Bayer AG, Bristol-Myers Squibb, and Medtronic. Dr Haverkamp
medical workers from all around the world have been reports grants from the German Federal Ministry of Education and Research
(BMBF) within the Medical Informatics Funding Scheme (MIRACUM; FKZ
joining campaigns urging people to stay home. Such 01ZZ1606H). MIRACUM is funded by the BMBF within the Medical Informatics
campaigns may contribute to decreasing admissions for Funding Scheme (FKZ 01ZZ1606H). Dr Brich reports personal fees and speaker
acute stroke, if not balanced by raising and sustaining honoraria from Bayer AG, Boehringer Ingelheim, and Daiichi Sankyo. The other
public awareness for stroke, in particular regarding the authors report no conflicts.

importance of timely diagnosis and treatment of acute

neurological symptoms, for example, through social
REFERENCES
media or respective information on-site. Patients should
1. Katan M, Luft A. Global burden of stroke. Semin Neurol. 2018;38:208–211.
be encouraged to seek medical care and be assured that doi: 10.1055/s-0038-1649503
hospitals make every effort to prevent contact to poten- 2. Bersano A, Pantoni L. On being a neurologist in italy at the time
tially infectious patients. All of this is paramount to avoid of the COVID-19 outbreak. Neurology. 2020;94:905–906.
serious healthcare and economic consequences result- doi: 10.1212/WNL.0000000000009508
3. Dashboard zu COVID-19 mit Fallzahlen nach Bundesland und Landkreis.
ing from undiagnosed and untreated strokes.
Downloaded from http://ahajournals.org by on October 3, 2021

https://www.rki.re/de/content/infaz/n/neuartiges_coronavirus/ncov.html.
Robert Koch Institute. Accessed April 20, 2020.
4. Klok FA, Kruip M, van der Meer NJM, Arbous MS, Gommers D, Kant KM,
ARTICLE INFORMATION Kaptein FHJ,van Paassen J,Stals MAM,Huisman MV, et al. Incidence of throm-
Received April 28, 2020; accepted May 12, 2020. botic complications in critically ill icu patients with COVID-19. Thromb Res.
2020;S0049-3848(20)30120-1. doi: 10.1016/j.thromres.2020.04.013
Affiliations 5. Dhand A, Luke D, Lang C, Tsiaklides M, Feske S, Lee JM. Social net-
Department of Neurology, Medical Faculty Mannheim, Heidelberg University, Ger- works and risk of delayed hospital arrival after acute stroke. Nat Commun.
many (C. Hoyer, A.E., M.P., K.S.). Department of Neurology, University Hospital 2019;10:1206. doi: 10.1038/s41467-019-09073-5

Stroke. 2020;51:2224–2227. DOI: 10.1161/STROKEAHA.120.030395 July 2020   2227