himpunan riset covid19

THE EARLY TRACHEOSTOMY IN CRITICALLY ILL COVID–19 PATIENTS

Dis Bima Purwaamidjaja1*, Mayang Indah Lestari2

1. Departement of Anesthesiology and Intensif Care RSPAD. Gatot Soebroto, Jakarta, Indonesia
2. Departement of Anesthesiology and Intensif Care Faculty of Medicine Universitas Sriwijaya– RSUP.

Dr. Moh. Hoesin, Palembang, Indonesia
*corresponding author

ABSTRACT
The main reason of critically ill COVID–19 patients being treated in intensive care unit is respiratory
failure which may require invasive mechanical ventilation support. Duration of this support tendsto be
long which may put patients into risk of complications such as increased mortality, weaning difficulties,
ventilator associated pneumonia (VAP), sedation requirement, and tracheal stenosis. Moreover,
COVID–19 condition is also aggravated by the large amounts production of thick secretions and may clot
endotracheal tube (ETT) causing oxygenation and ventilation problems. Tracheostomy is an alternative
airway management in critically ill COVID–19 patients. This technique provides benefitssuch as reducing
deadspace and resistance of the airways, work of breathing, sedation requirements, and injury to
oropharyng and laryng. Tracheostomy is safe and easy to be care, facilitates drainage of secretions,
improves oral nutrition, provides comfort and better communication. Percutaneous dilatational
tracheostomy is a safe procedure and can be performed bedside. Patients should have good prognosis
and get more benefits from tracheostomy since it’s an aerosol generating procedure transmits the
diseases into health workers This literature review will discuss about implications of COVID-19, choice
of tracheostomy and benefits of early tracheostomy in critically ill patients.

Keywords: COVID–19; critically ill; percutaneous dilatational tracheostomy; early tracheostomy

PENDAHULUAN Keputusan untuk penggunaan trakeostomi juga
Mayoritas pasien coronavirus disease 2019 bergantung dengan kondisi penyebab dan
(COVID–19) memiliki gejala infeksi sistem tingkat keparahan gagal napas pasien. 8,9Waktu
pernapasan yang sangat heterogen mulai dari untuk melakukan trakeostomi masih menjadi
gejala minimal hingga hipoksia berat akibat kontroversi. Belum ada pedoman yang
acute respiratory distress syndrome (ARDS).1 menyatakan waktu optimal untuk prosedur
Pandemi COVID–19 menyebabkan banyak trakeostomi. Berdasarkan waktu, trakeostomi
pasien mengalami sakit kritis dan sebanyak 5% dibagi menjadi trakeostomi dini dan lambat.
memerlukan perawatan intensif dan ventilasi Hingga saat ini belum ada kategori yang jelas
mekanis.2,3 Prosedur ventilasi mekanis yang terhadap definisi trakeostomi dini dan
paling sering dilakukan di ruangan intensive trakeostomi lambat.10 Beberapa studi pada
care units (ICU) ini mengharuskan pemasangan pasien ICU menunjukkan trakeostomi dini dapat
endotracheal tube (ETT) atau intubasi. mengurangi durasi ventilasi mekanis, lama
rawat ICU, mengurangi insiden VAP, dan
Namun, pasien yang diintubasi menyebabkan mengurangi mortalitas pasien kritis. Akan tetapi
kesulitan dalam perawatan oral, keterbatasan ada beberapa telaah artikel yang tidak
komunikasi, kesulitan pemberian nutrisi, menyetujui manfaat dari trakeostomi dini.7,9,11
kekurangnyamanan dan lain–lain. Ventilasi
mekanis dan ETT yang dipakai cenderung lama Ketika pandemi COVID–19 menyebar ke Italia
sehingga dapat meningkatkan risiko mortalitas, dan Spanyol, ruangan ICU menjadi sangat
ventilator–associated pneumonia (VAP), penuh dengan pasien kritis dan banyak
kesulitan penyapihan (weaning), dan perawatan membutuhkan trakeostomi. Trakeostomi juga
ICU yang lama.4 Data yang ada menunjukkan termasuk prosedur yang menimbulkan
pasien gagal napas akibat COVID–19 aerosolisasi meskipun dengan menggunakan
membutuhkan ventilasi mekanis kurang lebih 18 alat pelindung diri (APD) yang memadai.2 Peran
hari.5,6 Untuk mengantisipasi ventilasi mekanis trakeostomi pada pasien COVID–19 masih
yang lama, klinisi sering mempertimbangkan belum jelas. Trakeostomi dini pada pasien
prosedur trakeostomi.7 Trakeostomi dapat COVID–19 mungkin memiliki potensi manfaat.
mengurangi komplikasi intubasi lama seperti 5,6 Belum ada indikasi pasti terkait indikasi dan
VAP, sinusitis, stenosis trakea, mempercepat waktu trakeostomi pada pasien COVID–19.
weaning ventilator, mempercepat lepas rawat Indikasi untuk trakeostomi cenderung mengarah
ICU dan meningkatkan angka kesembuhan.4,7 pada ketersediaan sarana prasarana seperti
ventilator, ruang ICU, sedasi dan faktor
Trakeostomi umumnya diindikasikan pada pasien.6,12,13 Trakeostomi dapat menjadi salah
ventilasi lama atau pada pasien yang diprediksi satu strategi klinis dalam manajemen epidemi
membutuhkan ventilasi mekanis 10 hari atau berhubungan dengan gagal napas selama abad
lebih.4 Trakeostomi biasanya dilakukan pada ke-20 seperti poliomielitis dan difteri.2
hari ke-10 hingga ke-14 selama intubasi.
besar sitokin proinflamasi (IFN–α, IFN–γ, IL–1β,
IMPLIKASI COVID–19 PADA PASIEN IL– 6, IL–12, IL–18, IL–33, TNF–α,TGFβ, dll)
KRITIS dan kemokin (CCL2, CCL3, CCL5, CXCL8,
Sebanyak 5% pasien membutuhkan perawatan CXCL9, CXCL10, dll) akibat sel efektor dari
intensif atau ventilasi mekanis.14 Penyebab infeksi SARS– CoV2. Badai sitokin akan
kematian utama dari COVID–19 adalah ARDS. memicu sistem imun menyerang tubuh
Acute respiratory distress syndrome adalah menyebabkan ARDS, gagal napas dan
kondisi imunopatologis akibat infeksi SARS– kegagalan organ multipel sehingga
CoV2. Mekanisme utama penyebab ARDS membutuhkan ventilasi mekanis dan bahkan
adalah badai sitokin, suatu respons inflamasi dapat menyebabkan kematian pada kasus berat
sistemik tidak terkontrol dari pelepasan sejumlah pasien COVID–19.15

Selain menyerang sistem respirasi, COVID–19 trakeostomi yaitu surgical tracheostomy (ST)
diduga dapat menyerang sistem saraf pusat
(SSP). Rumah sakit di Beijing melaporkan dan percutaneous dilatational tracheostomy
adanya ensefalitis virus disebabkan serangan
coronavirus ke SSP. Hampir 40% pasien (PDT). Surgical tracheostomy meliputi diseksi
COVID–19 mengalami nyeri kepala, gangguan
kesadaran dan disfungsi otak lainnya. Badai jaringan pratrakea dan memasukan kanul
sitokin dan gangguan koagulasi juga dapat
meningkatkan risiko penyakit cerebrovaskular.16 trakeostomi dengan melihat trakea secara
Coronavirus menginfeksi neuron pada batang
otak yang berhubungan dengan kontrol langsung. PDT dilakukan dengan memasukkan
kardiorespirasi. Kerusakan area ini dapat
memperparah atau menyebabkan gagal kanul trakea dengan melakukan diseksi tumpul
napas.17 Gangguan pada kesadaran dan kontrol
pernapasan ini dapat membuat pasien COVID– pada jaringan pratrakea menggunakan teknik
19 membutuhkan ventilasi mekanis yang lama.
Data sebelumnya menunjukkan pasien gagal Seldinger.8,10 Trakeostomi adalah prosedur
napas akibat COVID–19 membutuhkan ventilasi
mekanis dengan durasi rata–rata 18 hari.5,6 invasif dengan komplikasi seperti
Durasi ventilasi mekanis yang lama ini akan
meningkatkan kebutuhan sedasi dan sering perdarahan,ulserasi, parut, gangguan kosmetik,
diberikan beragam jenis obat sedatif, hal ini
dapat menimbulkan keterbatasan persediaan infeksi, emfisema subkutan, pneumotoraks dan
pada obat pasien kritis dimasa pandemi
sehingga dibutuhkan solusi untuk mengurangi stenosis trakea. Akan tetapi, dibandingkan
kebutxuhan sedasi pada pasien COVID–19
yang membutuhkan ventilasi mekanis.18 dengan intubasi ETT jangka panjang,
Sebagai tambahan, biopsi dari pasien COVID–
19 memperlihatkan eksudat fibromiksoid dan trakeostomi memiliki beberapa keuntungan
pembentukan sumbatan sekret mukus yang
tebal. Otopsi pertama menemukan banyak seperti lebih sedikit dead space jalan napas dan
jumlah sekret kental yang keluar dari alveoli.
Temuan ini dapat memberikan implikasi resistensi jalan napas yang rendah, mengurangi
terhadap terapi klinis. Bila komponen mukus
tidak dibersihkan, maka pemberian oksigen work of breathing, meminimalkan lesi orofaring
tidak akan optimal dan dapat meningkatkan
hipoksia pada pasien.19–21 dan laring, meningkatkan nutrisi oral,

TRAKEOSTOMI kenyamanan pasien, komunikasi lebih baik,
Sebanyak 8–13% pasien rawat ICU
membutuhkan ventilasi mekanis dilanjutkan perawatan yang lebih mudah dan aman, dan
dengan trakeostomi.2 Indikasi utama trakeostomi
adalah pada pasien yang membutuhkan mengurangi sekret jalan napas.9–
ventilasi mekanis lama, akses untuk mengurangi
sekret jalan napas, obstruksi jalan napas atas 11,22Percutaneous dilatational tracheostomy
dan mengurangi ruang rugi (dead space) serta
untuk memfasilitasi weaning ventilator.8 adalah prosedur standar manajemen jalan
Trakeostomi adalah prosedur membuka dinding
anterior trakea dan diikuti dengan fiksasi trakea napas untuk pasien ICU yang membutuhkan
terhadap kulit di leher. Terdapat dua teknik
ventilasi jangka panjang. Teknik ini memiliki

beberapa keunggulan seperti prosedur yang

dapat dilakukan di bedside, metode yang aman

dengan angka mortalitas rendah dan memiliki

hasil kosmetik yg lebih baik setelah perawatan.

Indikasi PDT di ICU antara lain untuk

memfasilitasi kesulitan weaning, meningkatkan

higiene trakeobronkial, melindungi risiko

aspirasi, mengantisipasi ventilasi lama dan

meminimalkan kebutuhan sedasi. Meskipun

demikian, prosedur ini juga memiliki kelemahan

seperti membutuhkan operator

berpengalaman.23–25

TRAKEOSTOMI PADA PASIEN KRITIS
COVID–19
Peran trakeostomi pada pasien COVID–19
masih belum diketahui secara jelas.
Trakeostomi dini pada COVID–19 mungkin
berpotensi memiliki manfaat terhadap pasien.5,6
Belum ada indikasi jelas trakeostomi pada
pasien COVID–19.Indikasi trakeostomi mungkin
berdasarkan ketersediaan sarana–prasarana,
seperti ketersediaan ventilator, bed ICU dan
ketersediaan obat sedasi. Trakeostomi

tergolong suatu prosedur menimbulkan aerosol Belum ada teknik trakeostomi yang
yang berimplikasi terhadap transmisi infeksi direkomendasikan, teknik PDT dan ST dapat
kepada petugas kesehatan.6,12,13 Risiko digunakan. Surgical tracheostomy diindikasikan
prosedur ini harus dipertimbangkan mengingat pada pasien obesitas, leher pendek dan riwayat
terdapat risiko pasien terhadap intubasi lama, hipertrofi kelenjar tiroid.12 Teknik PDT meliputi
stenosis trakea dan kerusakan mukosa.26 manipulasi jalan napas yang luas seperti
Trakeostomi adalah salah satu metode untuk bronkoskopi dan/atau dilatasi pada trakea.
mempercepat lepas rawat ICU dan pindah ke Pasien juga membutuhkan koneksi dan
ruang rawat biasa sehingga dapat menjadi pemutusan koneksi berulang dari sirkuit
solusi penuhnya kapasitas ruang ICU yang saat ventilator. Sehingga PDT memiliki risiko
ini dialami saat pandemi COVID-19.12 aerosolisasi lebih tinggi secara teoritis. Teknik
Trakeostomi juga memungkinkan manajemen ST lebih digunakan dibanding PDT saat wabah
sekret yang lebih baik karena kemudahan saat SARS. Akan tetapi, teknik PDT telah sangat
suction dan kemampuan untuk mengganti maju dan belum ada data menunjukkan
kanula sehingga masalah banyaknya sekret kelebihan satu teknik terhadap teknik lainnya
pada pasien COVID-19 bisa diatasi. dalam hal mengurangi risiko transmisi
Penggunaan obat paralisis dan sedasi juga penyakit.2,27 Prosedur PDT dapat
dapat dikurangi pada penggunaan trakeostomi. dipertimbangkan pada pasien dengan anatomi
Manfaat potensial lainnya adalah menurunkan yang memungkinkan. Penggunaan
insidensi stenosis trakea serta menurunkan perlengkapan PDT yang sekali pakai lebih
resistensi jalan napas. Tidak kalah pentingnya, direkomendasikan.28 Teknik ini juga dapat
trakeostomi juga memudahkan untuk melakukan meminimalisir perpindahan pasien sehingga
terapi rehabilitasi respirasi, fisik dan pasien meminimalkan penyebaran transmisi. Angel
dapat duduk sehingga mempercepat fase dkk. Menerapkan teknik PDT baru dengan
penyembuhan.5,6 Terlepas dari segala manfaat, menempatkan bronkoskopi disamping ETT,
terdapat risiko timbulnya aerosolisasi ketika bukan didalamnya. Hal ini memberikan mitigasi
pasien batuk atau sedang dilakukan suction.6 risiko aerosolisasi virus selama prosedur.
Meskipun trakeostomi memiliki manfaat pada Seluruh tenaga medis yang berpartisipasi
pasien, prosedur ini tidak direkomendasikan dengan teknik PDT ini tidak terinfeksi
pada pasien yang masih membutuhkan fraksi coronavirus yang dibuktikan dengan negatif
oksigen tinggi, kebutuhan ventilator tinggi dan pada saat pemeriksaan dan tidak ada gejala
membutuhkan posisi prone. Pasien dengan COVID-19.5 Sejalan dengan hasil diatas,
trakeostomi dapat dilakukan posisi prone, penelitian dari Takhar dkk. melakukan PDT
namun jalan napas tidak dapat dilihat secara pada 51 pasien COVID-19. Mereka
jelas sehingga berisiko pergeseran posisi dan menggunakan PDT menggunakan ultrasound
kerusakan akibat tekanan. McGrath dkk. dan bronkoskopi untuk mengurangi risiko
menyarankan penggunaan trakeostomi ditunda aerosolisasi dan tidak ada tenaga klinis yang
hingga 10 hari penggunaan ventilasi mekanis terbukti PCR positif terhadap COVID-19.29
dan hanya pada pasien yang menunjukkan
gejala perbaikan klinis.2 Lokasi melakukan Berikut beberapa rekomendasi pertimbangan
prosedur sebaiknya dilakukan secara bedside di langkah mengurangi aerosolisasi dalam
ruangan ICU untuk meminimalkan transportasi melakukan PDT :30–32
pasien yang tidak dibutuhkan. 1. Penggunaan bronkoskopi tidak selalu
Trakeostomi sebaiknya dilakukan didalam
ruangan tekanan negatif untuk meminimalkan dibutuhkan, tetapi penggunaan mount
kontaminasi. Operator sebaiknya menggunakan kateter (konektor tube fleksibel) dengan
APD yang memadai meliputi pelindung wajah, port tertutup pada bronkoskopi dapat
powered air–purifying respirators (PAPRs), meminimalisir aerosolisasi
google, gown, dan sarung tangan. Suction 2. Prosedur dilakukan dalam sedasi dalam
dengan sistem tertutup lebih disarankan.6,27,28 dan blokade neuromuskular penuh. Sedasi
disarankan mencapai Richmond agitation

sedation scale -2 hingga +1. Penggunaan percobaan apnea tidak dapat menggantikan
blokade neuromuskular dalam mencapai sepenuhnya keputusan klinis multidisiplin
train-of-four ratio 0 sangat berdasarkan risiko dan manfaat terhadap
direkomendasikan untuk mencegah trakeostomi. 2
pergerakan pasien dan batuk.
3. Langkah pertama untuk mengempiskan MANFAAT TRAKEOSTOMI DINI
cuff ETT dan menarik ETT dengan Waktu dilakukannya trakeostomi mungkin
panduan laringoskop hingga cuff terlihat memiliki pengaruh terhadap kondisi pasien
pada ketinggian pita suara. kritis.9 Hingga saat ini belum ada literatur
Direkomendasikan untuk mengembangkan yang mendefinisikan trakeostomi dini atau
cuff untuk memastikan tidak ada kebocoran tertunda. Sebuah meta–analisis dari Huang
saat prosedur dkk, mendefinisikan trakeostomi dini adalah
4. Direkomendasikan melakukan klem trakeostomi yang dilakukan 48 jam hingga 3
terhadap ETT dan menghentikan ventilasi minggu setelah dilakukan intubasi.9 Beberapa
(pada akhir ekspirasi) selama langkah penting penelitian lain menyatakan trakeostomi dini
yang dapat meningkatkan risiko aerosolisasi: apabila dilakukan trakeostomi dalam 7 hari dan
ganti mount kateter, reposisi cuff ETT pada ada penelitian lain yang mengatakan dalam 10
ketinggian pita suara, dan buang dilator rhino hari setelah dilakukan pemasangan ETT.10,33,34
besar. Saat ini, mayoritas klinisi mendefinisikan
5. Lokasi pungsi trakea harus dilapisi dengan trakeostomi dini dilakukan dalam 1 hingga 2
swab selama prosedur untuk mengurangi minggu setelah intubasi. Oleh karena definisi
penyebaran aerosol. yang inkonsisten ini, hasil penelitian sejauh ini
sangat bervariasi dan belum dapat digunakan
Paralisis dengan menggunakan obat sebagai referensi.8,9
penghambat neuromuskular dapat mengurangi
pergerakan pasien dan batuk, tetapi takifilaksis Penelitian retrospektif dari Taiwan mendapatkan
akibat respons obat dapat terjadi pada pasien trakeostomi setelah 21 hari intubasi
kritis. Monitoring neuromuscular penting untuk berhubungan dengan tingkat kesulitan weaning,
memastikan paralisis adekuat selama prosedur mortalitas ICU dan durasi rawat yang lebih tinggi.
tra keostomi.2 Menghentikan ventilasi ketika Menunda trakeostomi atau intubasi lama dapat
memasukkan selang trakeostomi dapat merusak lapisan pelindung bronkus dan higine
meminimalisir penyebaran aerosol. Manuver bronkial, serta meningkatkan risiko kolonisasi
ajuvan seperti menempatkan cuff ETT dibawah bakteri.35 Sebuah review dari Cochrane tahun
daerah trakeostomi pada ST dapat meminimalisir 2015 dari 8 penelitian randomized control trial
durasi apnea. Apnea dapat menimbulkan (RCT) menyimpulkan bahwa tingkat mortalitas
hipoksia cepat pada pasien kritis dan dependen pada jangka pengawasan terpanjang pada
terhadap ventilator, sehingga disarankan kelompok trakeostomi dini lebih rendah, namun
melakukan preoksigenasi, diikuti dengan untuk mortalitas dalam 28 hari, 60 hari, 90 hari
percobaan apnea di ICU dengan FiO2 1.0 dan satu atau dua tahun tidak ada hubungan
signifikan. Empat penelitian juga menemukan
dan PEEP 5 cm H2O pada pasien dengan trakeostomi dini dapat mengurangi durasi
penggunaan ventilasi mekanis dan lama rawat
posisi supinasi sebelum trakeostomi. ICU.10,36 Trakeostomi dini juga dapat membantu
Desaturasi cepat pada percobaan ini dapat rehabilitasi aktif, meningkatkan kekuatan otot,
memprediksi respons yang mirip ketika fungsi fisik dan kualitas hidup.7
memasang trakeostomi dan trakeostomi dapat
ditunda. Kemampuan untuk mentoleransi trakeostomi dini (<10 hari intubasi) dapat
mengurangi biaya rumah sakit selama perawatan
Sebuah telaah sistematis oleh Meng dkk. di ICU.38 Penelitian terhadap pasien cedera kepala
Dan Hosokawa dkk. mendapatkan trakeostomi yang membutuhkan trakeostomi oleh Robba
dini juga berpotensi dapat mengurangi durasi
sedasi.11,37 Herrit dkk. juga menemukan

dkk, mendapatkan bahwa trakeostomi dini Berlawanan dengan hasil diatas, beberapa
(≤7 hari admisi) dapat mengurangi lama rawat penelitian dan review yang menyatakan
rumah sakit dan perbaikan fungsi neurologis trakeostomi dini tidak secara signifikan
pasien. Keterlambatan 2 hari trakeostomi dapat menurunkan mortalitas, risiko VAP, durasi
meningkatkan lama rawat di ICU sebanyak 1 hari ventilasi mekanis dan lama rawat ICU.4,8,9,37
dan lama rawat rumah sakit selama 2 hari.39 Hal Perbedaan ini disebabkan karena heterogenitas
ini menunjukkan bahwa modifikasi pada waktu dari definisi trakeostomi dini sehingga
trakeostomi dapat memperbaiki hasil perawatan memengaruhi hasil penelitian.
pada pasien.

Gambar1. Pertimbangan trakeostomi setelah intubasi pada pasien COVID-19
Keterangan: Jendela Trakeostomi 10-21 hari setelah intubasi. ICU = intensive care unit (telah
diterjemahkan).2

TRAKEOSTOMI DINI PADA PASIEN COVID- Menunda trakeostomi pada pasien COVID–19
19 dapat mengurangi risiko pada tenaga medis,
Waktu terbaik melakukan trakeostomi masih namun perlu dipertimbangkan durasi yang
dalam kontroversi dan butuh penelitian lebih lama dari intubasi translaring, ventilasimekanis,
lanjut. Sebuah panduan dari COVID–19 sedasi dan lama rawat ICU juga dapat
tracheostomy task force dari Universitas menimbulkan komplikasi. Trakeostomi dini dapat
Pennsylvania dan rekomendasi dari New York dipertimbangkan untuk memperbaiki kebersihan
Head and Neck Society merekomendasikan atau pembersihan sekret, mengurangi kebutuhan
trakeostomi dapat digunakan pada pasien sedasi, mengurangi infeksi dan inflamasi, pada
COVID–19 dengan intubasi lebih dari 21 hari kondisi mengancam nyawa, dan meningkatkan
untuk menghindari risiko paparan pada tenaga prognosis pasien. Pertimbangan terhadap waktu
medis.26,28 Bukti terbaru saat ini menunjukkan terbaik melakukan prosedur trakeostomi dapat
sebaiknya menunda trakeostomi paling tidak 14 dilihat pada gambar 1. Hindari trakeostomi pada
hari setelah intubasi.30 pasien dengan risiko mortalitas tinggi dan hindari

juga trakeostomi emergensi bila memungkinkan trakeostomi dini yang lebih banyak, disarankan
untuk meminimalisir risiko transmisi penyakit. untuk melakukan trakeostomi secepat mungkin
Trakeostomi pada pasien COVID–19 sebaiknya kecuali pada pasien dengan prognosis jelek dan
dipertimbangkan kasus per kasus dan belum dapat membahayakan tenaga medis.
bisa digeneralisir.2,6,26,28,30 Melihat manfaat

SIMPULAN 2020;6736(20):19–20.
Pasien COVID–19 cenderung membutuhkan
ventilasi mekanis dalam jangka panjang. Ventilasi 4. Koch T, Hecker B, Hecker A, Brenck F,
mekanis jangka panjang dapat menimbulkan Preuß
beberapa komplikasi. Trakeostomi memiliki
banyak manfaat pada pasien yang membutuhkan M. Early tracheostomy decreases ventilationtime
ventilasi mekanis jangka panjang seperti but has no impact on mortality of intensive
COVID–19. Perhatian terhadap produksi aerosol care patients : a randomized study.
harus dilakukan untuk pencegahan transmisi Springer. 2012;397:1001–8.
ke tenaga medis dengan menggunakan teknik
dan APD yang tepat. Trakeostomi dini berisiko 5. Angel L, Kon ZN, Chang SH, Rafeq S,
terjadinya penularan terhadap tenaga medis Shekar PS, Mitzman B, et al. Novel
namun trakeostomi dini dapat memperbaiki percutaneous tracheostomy for critically ill
higiene atau pembersihan sekret, mengurangi patients with COVID-19. Ann Thorac Surg.
kebutuhan sedasi, mengurangi infeksi dan 2020;1–17.
inflamasi, memperbaiki fungsi neurologis,
mengurangi lama rawat dan meningkatkan 6. Heyd CP, Desiato VM, Nguyen SA, Rourke
prognosis pasien. PDT dapat meminimalisir AKO, Clemmens CS, Awad MI, et al.
risiko penularan dalam rumah sakit karena Tracheostomy protocols during COVID-19
dapat dilakukan secara bedside. Trakeostomi pandemic. Head Neck. 2020;42(6):1–18.
dini dengan teknik PDT lebih disarankan karena
memiliki lebih banyak manfaat dengan tetap 7. Mccredie VA, Adhikari NKJ. Early
mempertimbangkan risiko transmisi virus tracheostomy in critically ill patients : still too
kepada tenaga medis dan prognosis pasien. fast. Lancet Respir. 2015;3(2):95–6.

KONFLIK KEPENTINGAN 8. Raimondi N, Vial MR, Calleja J, Quintero
Penulis menyatakan tidak ada konflik kepentingan A, Cortés A, Celis E, et al. Evidence-based
dalam penulisan artikel ini. guidelines for the use of tracheostomy in
critically ill patients. J Crit Care.
DAFTAR PUSTAKA 2017;38:304–18.

1. WHO. World Health Organization. Global 9. Huang H, Li Y, Ariani F, Chen X, Lin J.
Timing of tracheostomy in critically ill
Yuki K, Fujiogi M, Koutsogiannaki S. COVID- patients : a meta- analysis. PLoS One.
2014;9(3):1–11.
19pathophysiology: a review. Clin Immunol.
10. Bng A, Rb A, Saconato H, Án A, Valente O.
2020;215(108427):1–7. Early versus late tracheostomy for critically
ill patients ( review ). Cochrane Collab.
2. McGrath BA, Brenner MJ, Warrillow SJ, 2018;(12):1–55.

Pandian V, Arora A, Cameron TS, et al. 11. Hosokawa K, Nishimura M, Egi M,
Vincent
Tracheostomy in the COVID-19 era: global
J. Timing of tracheotomy in ICU patients : a
and multidisciplinary guidance. Lancet systematic review of randomized controlled
trials. Crit Care. 2015;19(424):1–12.
Respir. 2020;2600(20):1–8.
12. Mattioli F, Fermi M, Ghirelli M, Molteni G,
3. Baker T, Schell CO, Petersen DB, Sawe H, Sgarbi N, Bertellini E, et al.
Tracheostomyin the COVID - 19 pandemic.
KhalidK, Mndolo S, et al. Essential care of Eur Arch Oto- Rhino-Laryngology. 2020;1–
3.
critical illness must not be forgotten in the
13. Pichi B, Mazzola F, Bonsembiante A,
COVID-19 pandemic. Lancet.

Petruzzi G, Zocchi J, Moretto S, et al. 24. Cho Y. Percutaneous dilatational

CORONA-steps for tracheotomy in COVID- tracheostomy. Tuberc Respir Dis (Seoul).

19 patients: a staff-safe method for airway 2012;72(3):261–74.

management. Oral Oncol. 2020;105:1–3. 25. Suzuki Y, Suzuki T, Yamamoto Y,

14. Baker T, Schell CO, Petersen DB, Sawe H, Teshigawara A, Okuda J, Suhara T, et

KhalidK, Mndolo S, et al. Essential care of al. Evaluationof the safety of percutaneous

critical illness must not be forgotten in the dilational tracheostomy compared with

COVID-19 pandemic. Lancet. surgical tracheostomy in the intensive care

2020;395:1253–4. unit. Crit Care Res Pract. 2019;2019:1–5.

15. Li X, Geng M, Peng Y, Meng L, Lu S. 26. Miles BA, Schiff B, Ganly I, Ow T, Cohen E,

Molecular immune pathogenesis and Genden E, et al. Tracheostomy during the

diagnosis of COVID-19. J Pharm Anal. COV- SARS-CoV-2 pandemic:

2020;10(2):102–8. recommendations from the New York Head

16. Wu Y, Xu X, Chen Z, Duan J, Hashimoto K, and Neck Society. Head Neck.

Yang 2020;42(6):1–14.

L. Nervous system involvement after 27. Tay JK, Khoo MLC, Loh WS. Surgical

infection with COVID-19 and other considerations for tracheostomy during the

coronaviruses. Brain, Behav Immun. COVID-19 pandemic lessons learned from

2020;30(40):1–5. the severe acute respiratory syndrome

17. Steardo L, Jr LS, Zorec R, Verkhratsky A. outbreak. JAMA Otolaryngol Neck Surg.

Neuroinfection may potentially contribute to 2020;1–2.

pathophysiology and clinical manifestations 28. Chao TN, Braslow BM, Martin ND, Chalian

of COVID-19. Acta Physiol. 2020;1–4. AA, Atkins JH, Haas AR, et al.

18. Madhok J, Mihm FG. Rethinking sedation Tracheostomy in ventilated patients with

during prolonged mechanical ventilation for COVID-19. Ann Surg. 2020;1–7.

COVID-19 respiratory failure. Anesth Analg. 29. Takhar A, Tornari C, Amin N, Wyncoll DL,

2020;1–2. Tricklebank S, Arora A, et al. Percutaneous

19. Xu Z, Shi L, Wang Y, Zhang J, Huang L, tracheostomy in COVID-19 pneumonitis

Zhang C, et al. Pathological findings of patients requiring prolonged mechanical

COVID-19 associated with acute ventilation: Initial experience in 51 patients

respiratory distress syndrome. Lancet and preliminary outcomes. Researchgate.

Respir. 2020;8(4):420–2. 2020;1–7.

20. Mao Y, Lin W, Wen J, Chen G. Clinical 30. Takhar A, Walker A, Tricklebank S, Wyncoll

and pathological characteristics of 2019 D, Hart N, Jacob T, et al. Recommendation

novel coronavirus disease ( COVID-19 ): a of a practical guideline for safe

systematic reviews. medRxiv. 2020;1–31. tracheostomy during the COVID - 19

21. Tian S, Hu W, Niu L, Liu H, Xu H, pandemic. Eur Arch Oto-Rhino-

Xiao S. Laryngology. 2020;1–12.

Pulmonary pathology of early phase 2019 novel 31. Gosling AF, Bose S, Gomez E, Cook C,

coronavirus (COVID-19) pneumonia in two Shaefi S, Leibowitz A. Perioperative

patients with lung cancer. J Thorac Oncol. considerations for tracheostomies in the era

2020;1–14. of COVID-19. Crit Care Resusc.

22. Hyde GA, Savage SA, Zarzaur BL, Hart- 2020;XXX(Xxx):1–9.Purwaamidjaja D. B.

hyde JE, Schaefer CB, Croce MA, et al. dkk.

Early tracheostomy in trauma patients 32. Payen J-F, Chanques G, Futier E, Velly L,

saves time and money. Injury. Jaber S, Constantin J. Sedation for critically

2015;46(1):110–4. ill patients with COVID-19:Which

23. Mehta C, Mehta Y. Percutaneous specificities?One size does not fit all. Anesth

tracheostomy. Ann Card Anaesth. Crit Care Pain Med. 2020;39(January):341–

2017;20(5):S19-25. 3.

33. Griffiths J, Barber VS, Morgan L, Young JD.
Systematic review and meta-analysis of
studies of the timing of tracheostomy in adult
patients undergoing artificial ventilation.
BMJ. 2005;1–5.

34. Wang F, Wu Y, Bo L, Lou J, Zhu J, Chen F,
et al. A systematic review and meta-
analysis of randomized controlled trials.
Chest. 2011;140(6):1456–65.

35. Hsu C, Chen K, Chang C, Jerng J, Yu C.
Timing of tracheostomy as a determinant of
weaning success in critically ill patients : a
retrospective study. Crit Care.
2005;9(1):46–52.

36. Keeping A. Early versus late tracheostomy
for critically ill patients : a clinical evidence
synopsis of a recent Cochrane Review.
Can J Respir Ther. 2016;52(1):27–8.

37. Meng L, Wang C, Li J, Zhang J. Early vs
late tracheostomy in critically ill patients : a
systematic review and meta-analysis. Clin
Respir J. 2016;684–92.

38. Herritt B, Chaudhuri D, Thavorn K, Kubelik
D, Kyeremanteng K. Early vs. late
tracheostomy in intensive care settings:
impact on ICU and hospital costs. J Crit
Care. 2017;44:285–8.

39. Robba C, Galimberti S, Graziano F,
Wiegers EJA, Lingsma HF, Iaquaniello
C, et al. Tracheostomy practice and
timingin traumatic brain - injured patients : a
CENTER - TBI study. Intensive Care Med.
2020;46(5):983–94.

MINIMAL OR NO TOUCH ELECTROCARDIOGRAPHY RECORDING
AND REMOTE HEART RHYTHM MONITORING DURING COVID-19

PANDEMIC ERA

Alexander Edo Tondas1,2, Rolando Agustian2, Moza Guyanto2
1 Biomedicine Doctoral Program, Faculty ofMedicine, Universitas Sriwijaya, Palembang, Indonesia
2 Cardiology & Vascular Medicine Department, Mohammad Hoesin General Hospital, Palembang,Indonesia

Koresponden:
Alexander Edo Tondas MD,
Cardiology & Vascular Medicine Department, Mohammad Hoesin General Hospital, Palembang,Indonesia
E-mail: [email protected]

ABSTRACT
At the end of year 2019, the world faced an outbreak of a highly virulent novel Coronavirus disease (COVID-
19), whichchanged the way physicians, including cardiologists, do their routine clinical practice. As distance limitation
and efficient use of personal protective devices must be employed to prevent the pandemic spreading, even
simple electrocardiogram (ECG) taking that involves directly placing electrode leads on a patient’s body may become
riskier. This review will discuss the possibility of minimal or no touch EKG using the latest wireless technologies,
beneficial in monitoring COVID-19 patients for cardiovascular problems or patients who seek cardiac care, but with
posing risk of concomitant COVID-19.

(Indonesian J Cardiol. 2020;41:133-141)

Keyword: COVID-19, SARS-CoV-2, physical distancing, electrocardiogram, heart rhythm monitoring

INTRODUCTION preventing the spread of infection are through
Coronavirus is a large group of viruses that regular hand washing, applying the ethics of
cause mild to severe respiratory tract disease. coughing and sneezing, and avoiding direct
There are at least two types of coronavirus that contact. Minimizing direct contact is currently
are previously known to cause severe applied through physical distancing policies. The
complaints such as Middle East Respiratory physical distancing policy recommended by the
Syndrome (MERS) and Severe Acute Centers for Disease Control and Prevention
Respiratory Syndrome (SARS). The virus (CDC) includes limiting a minimum distance of 2
related to coronavirus disease 2019 (COVID-19) meters from other people and avoiding crowds
is SARS-CoV-2, a new entity firstly identified in or groups, which is the range of potential droplet
the city of Wuhan, China on December 2019 release by a patient or people at risk by
and since then the pandemic has escalated on a coughing, talking or sneezing.4 Distance
global scale rapidly, with increasing mortality limitation has proven effective in reducing
rate. Based on scientific evidence, COVID-19 COVID-19 transmission rates, as evidenced by
can primarily be transmitted from human to the declining of new COVID-19 patient curve
human through close contact and respiratory model after the policy of distancing was carried
droplets.1 Currently, no airborne transmission out.5 A study from Harvard, estimated from a
was reported from an analysis of 75,465 patients computer modeling that recurrent wintertime
in China.2 The people most at risk of contracting outbreaks of SARS-CoV-2 will probably occur
this disease are people who are in close contact after the initial wave, and prolonged or
with COVID-19 patients, including those who intermittent physical distancing should be
treat COVID-19 patients.3 Recommendations for maintained possibly as late as the year 2022 to

hasten the acquisition of herd immunity towards experimental studies show chloroquine and its
this disease.6 Therefore, a radical change in derivatives can inhibit the process of viral
how we approach patients in medicine, receptor glycolization on cellular surfaces,
especially in cardiology should be anticipated. including ACE-2 receptors, so that it cannot bind
Currently there is limited data on the spesific to these ACE-2 receptors expressed in lung,
ECG changes associated with SARS-CoV-2. He heart, kidney, and intestine.12
et al. obtained a series of ECG variations in
COVID-19 patients, such as McGinn-White sign However, hydroxychloroquine has significant
(S1Q3T3), AV block, ST elevation change, cardiovascular side effects, that is prolongation
ventricular tachycardia, sinus tachycardia, to of the QT interval, related to abnormal
right bundle branch block. Furthermore, some repolarization process of the ventricular
patients with COVID-19 may present features myocardium.9 The normal corrected QT interval
similar to myocarditis7 or come with using Bezett’s formula in women is around 470
emergencies such acute coronary syndromes ms and 450 ms in men. QT interval prolongation
(ACS), requiring immediate electrocardiogram causes concern because of it’s relationship with
(ECG) taking. Likewise, some medications the occurrence of sudden cardiac death,
under investigation for COVID-19 treatment may especially fatal torsade de pointes, a
have cardiovascular complications.8 In these polymorphic ventricular arrhythmia.13 The
scenarios, minimal or no-touch technique in mechanism of hydroxychloroquine in inducing
ECG acquisition maybe beneficial to minimize prolongation of the QT interval is not fully
direct exposure to SARS-CoV-2. understood. Capel et al. In their 2015 study
showed that hydroxychloroquine has an
CARDIOVASCULAR IMPLICATIONS IN inhibitory effect on hyperpolarization-activated
THERAPEUTIC OP- TIONS FOR COVID-19 ion channels (also known as "funny current"
To date, there has been no truly globally agreed- channels, If) along with delayed rectifier
upon therapy as standard therapy for COVID-19. potassium currents. (IKr) and L-type calcium ion
However, there have been several drugs currents (ICal). The inhibitory effect of
proposed as alternative therapies in COVID-19 hydroxychloroquine on pacemaker cells is
patients. Hydroxychloroquine has been included known to cause a delay in depolarization speed
in the COVID-19 management guidelines issue and a decrease in heart rate.14 The use of
and led by the Indonesian Lung Doctors hydroxychloroquine, especially when
Association, in addition to other additional administered concomitantly with other QT
therapies such as azithromycin, vitamin C, prolonging drugs such as azithromycin, requires
oseltamivir, etc. The recommended use of close ECG monitoring. Some medical centers
hydroxychloroquine is 400 mg / 24 hours for 5 suggested recommendations for this vulnerable
days.9 Chloroquine and its derivatives allegedly group of patients: (1) to perform a basic ECG
have the ability to inhibit the replication of recording, (2) to withhold drug administration in
several intracellular microorganisms including patients with baseline QT prolongation (eg, QTc
coronaviruses in vitro. Chloroquine and ≥500 ms) or with known congenital long QT
hydroxychloroquine have a similar mechanism syndrome, (3) to monito rheart rhythm and QT
of action, that is increasing endosomal pH and intervals, as well as to stop the drug if the QTc is
interfering with the SARS-CoV-2 cellular at a value of ≥500 ms, and (4) to prevent other
receptor glycosylation process, and ultimately things that can aggravate the condition of QTc
inhibit the infection ability of the virus itself.10 prolongation, such as accompanying drugs or
Chloroquine also contains the enzyme quinone electrolyte abnormalities (Figure 1).15 In
reductase-2, which is involved in sialic acid cooperative and able patients, wearable and
biosynthesis (acidic monosaccharides of cell wireless technologies may assist physicians to
transmembrane proteins required for ligand record and monitor patient’s heart rhythm safely
recognition), that allows chloroquine to be used and remotely, thus minimizing risky close
as a broad-spectrum antiviral agent.11 Some encounters.

Figure 1. QTc evaluation flowchart in COVID-19 Patients treated with Chloroquine/ Azithromycin.16
Image used with permission from the respective authors.

WEARABLE TECHNOLOGIES FOR MINIMAL The Apple watch utilized a light sensor
OR NO TOUCHECG ACQUISITION technology, able to measure blood flow at
Several ECG recording methods utilize the latest certain intervals using the principle of
technological developments. The advancement photoplethysmography, and to detect changes in
of mobile phone features has played an heart rate regularity. The algorithm will measure
important role in the process of discovering new intervals between beats, called tachograms, and
ways to record ECGs. Apple Watch is a type of if there are 5 out of 6 intervals within 48 hours
smartwatch released by giant technology that are outside the normal value, the application
company, Apple, which has some of supportive will display a notification.20 In another study
features.17 Unlike the previous series, the latest conducted by Saghir et al, a 100% rhythm
4th generation Apple Watch features a built- in interpretation concordance was found between
application able to record and display a single lead I of Apple 4th series ECG and lead I of 12
lead ECG, thus allowing it to be used leads ECG. The study also found moderate to
independently without having to be paired with strong agreement in manual measurement of
an iPhone.18 The user simply attaches one heart rate and basic intervals between devices
finger to the digital crown clock and a closed represented by Bland-Altman plots, including
circuit will be formed which will produce an ECG QTc interval. 21 Based on this finding, it is likely
image(Figure 2). 19ECG recording technology on that the Apple Watch may
the Apple Watch series 4 has also received FDA
clearance for early detection of atrial fibrillation
(AF) for consumer device. A large study
involving approximately 419,000 research
subjects, the Apple Heart Study, showed that
Apple Watch has a high degree of accuracy in
the detection of AF, with diagnostic values that
are almost comparable to 12-leads ECG.

Figure 2. Single-lead ECG taking in Apple Watch Series 4

Figure 3. Simultaneous 6 leads ECG recording in AliveCor Kardia Mobile 6L by placing the device on
the knee or ankle.

have a role in the diagnosis of other heart demodulates it into a digital ECG tracing results
rhythm disturbance, however, the access to its which can be stored in cloud storage, mobile
ECG feature is currently restricted in Asia, storage, or directly sent to a particular email.24,25
including Indonesia . Another challenge in the Kardia Mobile has the sensitivity and specificity
utility of Apple Watch for routine clinical practice of 98% and 91.4-97% for detecting atrial
is affordability, especially in developing fibrillation, as stated by Lowres et al and Lau et
countries. Actually, Alivecor is the first company al in their study. 25–27 The up to date version of
to get FDA clearance for medical-device Kardia that has been released and received
accessory products, namely KardiaBand, the FDA clearance for monitoring QT duration in
original ECG recording device attached on early patients receiving medications that can cause
version of the Apple Watch. The recording life-threatening QT prolongation is Kardia Mobile
technique is done by placing a finger on a steel 6L, which is able to display six-leads ECG
plate located on a smartwatch strap.22,23 After images for 30 seconds on leads I, II, III, aVF,
the discontinuation of KardiaBand following the aVR, and aVL. Kardia Mobile 6L has 3 sensors,
release of Apple Watch series 4 with its own two of them are similar previous version, while
ECG recording hardware and software, the the other one on the bottom of the plate is meant
Kardia Mobile by AliveCor entered the market. to touch the left knee during recording (Figure
Kardia Mobile works with an external case or 3). 28 With a relatively large number of acquired
plate plus a downloadable application to the iOS leads compared to other monitoring devices,
or Android operating systems on smartphones. coupled with a very minimal touch and cableless
The metal sensors on the plate receive the measurement method, Kardia Mobile 6L can
cardiac electric signal which is then converted to become an attractive alternative as a remote
an ultrasound FV sound signal (18-24 kHz). ECG recording tool for COVID-19 patients.
Sequently, the application in the smartphone

Figure 4. Sanketlife device and steps to take 12-lead ECG recording

For more consequential cases such as acute reports can be saved in the cloud for future
coronary syndromes, attempts had been made reference or downloaded and instantly shared
to modify the AliveCor Heart Monitor in order to with the healthcare professional for immediate
acquire 12 leads recording in the ST- LEUIS clinical advice.30,31
Study, by attaching two jumper cables to the
device and sequentially connect one cable to Dyundi et al. studied ECG samples obtained
precordial ECG stickers and another to the left using Sanketlife from 1521 random participants
arm (VnL) and the right arm (VnR) as grounding. and found 15 types of ECG abnormalities from
where a comparison of ST elevation detection 324 samples (21.3%). This result demonstrated
was carried out on 12 lead of Alivecor Heart the applicability of the device as a diagnostic
Monitor and conventional 12 lead ECGs. test in detecting heart abnormalities from the
Despite the promising feasibility of using a general population.32 A comparison study
smartphone to obtain a “12-lead equivalent” between Sanketlife and standard 12-lead ECG
ECG recording, the ability for this technology to was conducted by Kumar et al with 100
be utilized autonomously by non-medical participants. They found that Sanketlife has
personnel still requires further assessment. 29 identified all major ECG abnormalities in high
accordance with standard 12-lead ECG with
Although not FDA approved, Sanketlife is an high sensitivity (98.15%) and specificity (100%).
Indian-manufactured, low cost, pocket-sized, It also has high positive predictive value (100%)
and leadless ECG monitor that can support up and negative predictive value (97.36%).30
to 12 leads. The device is mechanized by Sanketlife has been tested in the general
SanketLife app running on compatible iOS and population of India as a quite affordable and
Android phones that connect wirelessly via practical point-of-care ECG recording method33.
Bluetooth technology to the device. There are It have the ability to be used in the outpatient
two sensors on the front panel of the device for department by physicians, technicians, and even
extremity leads and one on the top to be used patients with reasonable accuracy, and can
as the chest lead sensor. The 12 lead ECG can prove to be useful in COVID pandemic era.31 But
be measured without jelly or any actual like every other transient ECG monitoring
electrode, but it must be obtained serially, and device, it has difficulty in identifying time-
not synchronously (Figure 4). The obtained ECG dependent atrial and ventricular premature beats

where continuous 24-hour monitoring is a better devices have only one ECG channel.37
method.30 Specifications of each mobile ECGs Unfortunately, the Zio patch and SEEQ MCT
are outlined in Table 1. never reached the Indonesian market. In
Indonesia, cardiologists are probably more
Ambulatory electrocardiographic (AECG) familiar with MyPatch device which offers
monitoring is the most widely used method to several patch sizes for adult, pediatrics and
detect cardiac arrhythmias in the outpatient neonates alike, or the Spyder device with its
ambulatory setting at a longer term. The most Liquid-ECG cloud data storing system. MyPatch
commonly used AECG is 24-hour Holter device is waterproof while the Spyder device
monitoring.34,35 However, conventional 24-hour has the option to be removed from its patch
Holter monitoring often fails to detect the culprit during shower and replaced to resume recording
arrhythmia in patients with symptomatic afterwards. The comparison of different popular
arrhythmia. In the investigation of patients with ambulatory ECG devices is described inTable 2.
palpitations, 24-hour Holter monitoring is
reported to have a diagnostic yield of 15% to CONCLUSION
39%.34 The Holter is bulky and some patients As the prevalence of COVID-19 increases
find it uncomfortable to wear. With advances in exponentially, patients presenting with
technology, miniaturization of such seemingly non-related medical problems may
instrumentations is progressing rapidly in expose health care providers to increased risk of
concert with the evolution of microelectronic contracting the disease. This mandates for a
circuits and wireless networking technologies. change in the way cardiologists perform their
Patch-based second generation AECG routine medical practice cardiology, including
monitoring devices are now available. These
patch-based devices have the capability to
transmit ECG data wirelessly and can record
AECG for longer periods compared to
conventional 24-hour Holter.35,36 In the COVID-
19 era, such devices can be a godsend for
medical staff to perform minimal touch ECG
monitoring, where less cables means less
hassle. The Zio patch and SEEQ Mobile Cardiac
Telemetry (MCT) are two of CE marked, FDA
approved, water- resistant, wireless patch-based
AECG available.37 Barret et al. compared the Zio
patch with conventional Holter monitor. They
found that the Zio patch detected significantly
more arrhythmia events than the Holter
monitor.34 Vanegas-Cadavid et al. used SEEQ
MCT to monitor 100 participants and showed
that extended cardiac monitoring using SEEQ
MCT was capable of detecting ECG abnormality
in 22% of the patients.38 Fung et al. compared
Zio patch, SEEQ MCT and the standard Holter
monitor. The Zio patch, SEEQ MCT, and Holter
have data storing capacity of 14 days, 7.5 days
(can be extended to 30 days), and 24-72 hours
respectively. Both patch-based devices are
water-resistant while the standard Holter monitor
is not. However, the Holter monitor has multiple
ECG channels (3-12) while both patch-based

simple ECG taking to heart rhythm monitoring. practice, especially cardiology in response to
Wireless devices and remote technologies may COVID-19.
help cardiologists to minimize direct contact
between health care personnel and COVID-19 CONFLICT OF INTERESt
patients as an integral step in limiting its spread None
and resource utilization.
PUBLICATION AGREEMENT
ACKNOWLEDGEMENT The authors of this article permit Indonesian
We express our deep respect to all medics and Journal of Cardiology to publish this article if this
paramedics who are fighting the pandemic in the article is accepted
frontlines and we hope this manuscript may
contribute in the new paradigm of medical

FUNDING Coronavirus Disease 2019 (COVID-19): A
Self funding.
Review. Jama. 2020;2019.
We the Authors of this manuscript declare that
we have received the proper permission doi:10.1001/jama.2020.6019
regarding the use of figures/images/illustrations
from their Original Authors 11.Awadhesh Kumar Singh, Akriti Singh,

REFERENCES Altamash Shaikh, Ritu Singh AM.
1. Aylward, Bruce (WHO); Liang W (PRC).
Chloroquine and hydroxychloroquine in the
Report of the WHO-China Joint Mission on
Coronavirus Disease 2019 (COVID-19). treatment of COVID-19 with or without
WHO-China Jt Mission Coronavirus Dis
2019. 2020. diabetes: A systematic search and a
2. Ong SWX, Tan YK, Chia PY, et al. Air,
Surface Environmental, and Personal narrative review with a special reference to
Protective Equipment Contamination by
Severe Acute Respiratory Syndrome India and other developing countries.
Coronavirus 2 (SARS-CoV-2) from a
Symptomatic Patient. JAMA - J Am Med Diabetes Metab Syndr Clin Res Rev.
Assoc. 2020. doi:10.1001/jama.2020.3227
3. Kementrian Kesehatan. Pedoman 2020;(January).doi:10.1016/j.
Pencegahan dan Pengendalian COVID-19.
4. Center for Disease Control and Prevention dsx.2020.03.011
(CDC). What You Need to Know about
coronavirus disease 2019 (COVID-19). 12.Devaux CA, Rolain J-M, Colson P, Raoult D.
Choice Rev Online. 2020.
doi:10.5860/choice.48-1502 New insights on the antiviral effects of
5. Ferguson NM, Laydon D, Nedjati-Gilani G, et
al. Impact of non-pharmaceutical chloroquine against coronavirus: what to
interventions (NPIs) to reduce COVID-19
mortality and healthcare demand. Imperial expect for COVID-19? Int J Antimicrob
AcUk.2020;(March):3-20.doi:10.25561/77482
6. Kissler SM, Tedijanto C, Goldstein E, Grad Agents. 2020;(December 2019):105938.
YH, Lipsitch M. Projecting the transmission
dynamics of SARS-CoV-2 through the post- doi:10.1016/j.ijantimicag.2020.105938
pandemic period. medRxiv. 2020.
doi:10.1101/2020.03.04.20031112 13.Chen CY, Wang FL, Lin CC. Chronic
7. Ma K-L, Liu Z-H, Cao C, et al. COVID-19
Myocarditis and Severity Factors An hydroxychloroquine use associated
Adult Cohort Study. medRxiv. 2020.
doi:10.1101/2020.03.19.20034124 with QT prolongation and refractory
8. Division of Viral Diseases - National Center
for Immunization and Respiratory Diseases. ventricular arrhythmia. Clin Toxicol.
Therapeutic Options for COVID-19 Patients.
Centers for Disease Control and Prevention. 2006;44(2):173-175.
9. IDAI TC-19. Protokol Tatalaksana COVID-19.
2020. doi:10.1080/15563650500514558
10.Sanders JM, Monogue ML, Jodlowski TZ,
Cutrell JB. Pharmacologic Treatments for 14.Capel RA, Herring N, Kalla M, et al.

Hydroxychloroquine reduces heart rate by

modulating the hyperpolarization-activated

currentIf:Novelelectrophysiologicalinsightsan

dtherapeutic potential. Hear Rhythm.

2015;12(10):2186-2194.

doi:10.1016/j.hrthm.2015.05.027

15.Roden DM, Harrington RA, Poppas A, Russo

AM. Considerations for Drug Interactions on

QTc in Exploratory COVID-19 (Coronavirus

Disease 2019) Treatment.

Circulation.2020;19:1-6.

doi:10.1161/CIRCULATIONAHA120047521

16.Mitra RL, Greenstein SA, Epstein LM. An

algorithm for managing QT prolongation in

Coronavirus Disease 2019 (COVID-19)

patients treated with either chloroquine or

hydroxychloroquine in conjunction with

azithromycin: Possible benefits of

intravenous lidocaine. Hear Case Reports.

2020;2019. doi:10.1016/j.hrcr.2020.03.016

17.Isakadze N, Martin SS. How useful is the

smartwatch ECG? Trends Cardiovasc Med.

2019. doi:10.1016/j. tcm.2019.10.010

18.Apple Watch Series 5 - Health & ECG

19.Apple Watch ECG app: What cardiologists lead ECG in the detection of cardiovascular
want you to know - CNET.
diseases in a tertiary care setting. Indian
20.Turakhia MP et al. Results Of A Large-scale,
App- based Study To Identify Atrial Pacing Electrophysiol J. 2020;20(2):54-59.
Fibrillation Using A Smartwatch: The Apple
Heart Study - CRTOnline. doi:10.1016/j.ipej.2019.12.011

21.Saghir N, Aggarwal A, Soneji N, Valencia V, 31.Bharath Raj Kidambi, Aayush Kumar Singal
Rodgers G, Kurian T. Apple Watch Series 4
Vs. 12-Lead Ecg: a Comparison of Manual SKR. Smart Phone Electrocardiogram –
Electrocardiographic Waveform Analysis.J
Am Coll Cardiol. Bridging the Gap. J Pract Cardiovasc Sci.
2020;75(11):378.doi:10.1016/s0735-
1097(20)31005-6 2019. doi:10.4103/jpcs. jpcs_72_18

22.FDA clears AliveCor’s Kardiaband as the first 32.Esha Dyundi, Prashant Gupta, Robin
medical device accessory for the Apple
Watch | TechCrunch. Choudhary, KK Aggarwal, Sanchita Sharma

23.KardiaBand - EKG Band for Your Apple PB. Evaluation of the Prevalence of
Watch® | AliveCor – AliveCor, Inc.
Cardiovascular Disease in Urban Delhi Using
24.Walker A, Muhlestein J. Smartphone
electrocardiogram monitoring: current a Handheld ECG. 2019;30(4):314- 317.
perspectives. Adv Heal Care Technol.
2018;Volume4:15-24. 33.Dyundi E, Gupta P, Choudhary R. Identifying
doi:10.2147/ahct.s138445
the Prevalence of the Life-threatening Atrial
25.Bansal A, Joshi R. Portable out-of-hospital
electrocardiography: A review of current Fibrillation Using a Smartphone-Based
technologies.J Arrhythmia. 2018;34(2):129-
138. doi:10.1002/joa3.12035 Wireless Electrocardiography Device: An

26.Lau JK, Lowres N, Neubeck L, et al. IPhone Observational Study. J Pract Cardiovasc Sci.
ECG application for community screening to
detect silent atrial fibrillation: A novel 2019. doi:10.4103/ jpcs.jpcs_42_19
technology to prevent stroke. Int J Cardiol.
2013. doi:10.1016/j. ijcard.2013.01.220 34.Barrett PM, Komatireddy R, Haaser S, et al.

27.Lowres N, Neubeck L, Salkeld G, et al. Comparison of 24-hour Holter monitoring
Feasibility and cost-effectiveness of stroke
prevention through community screening for with 14-day novel adhesive patch
atrial fibrillation using iPhone ECG in
pharmacies: The SEARCH-AF study. electrocardiographic monitoring. Am J Med.
Thromb Haemost. 2014. doi:10.1160/TH14-
03-0231 2014;127(1):95.e11-95. e17.

28.KardiaMobile 6L | Portable 6-Lead EKG doi:10.1016/j.amjmed.2013.10.003
Device – AliveCor, Inc.
35.Karunadas CP, Mathew C. Comparison of
29.Muhlestein JB, Le V, Albert D, et al.
Smartphone ECG for evaluation of STEMI: arrhythmia detection by conventional Holter
Results of the ST LEUIS Pilot Study. J
Electrocardiol. 2015;48(2):249- 259. and a novel ambulatory ECG system using
doi:10.1016/j.jelectrocard.2014.11.005
patch and Android App, over 24 h period.
30.Kumar S, Nagesh CM, Singh M, et al.
Assessment of diagnostic accuracy of Indian Pacing Electrophysiol J.
SanketLife – A wireless, pocket-sized ECG
biosensor, in comparison to standard 12 2020;20(2):49-53.doi:10.1016/j.

ipej.2019.12.013

36.Steinberg JS, Varma N, Cygankiewicz I, et

al. 2017 ISHNE-HRS expert consensus

statement on ambulatory ECG and external

cardiac monitoring/ telemetry. Hear

Rhythm.2017;14(7):e55-e96.

doi:10.1016/j.hrthm.2017.03.038

37.Fung E, Järvelin MR, Doshi RN, et al.

Electrocardiographic patch devices and

contemporary wireless cardiac monitoring.

Front Physiol. 2015;6(MAY):1-10.

doi:10.3389/ fphys.2015.00149

38.Vanegas-Cadavid D, Valderrama-Barbosa Z,

Ibatá- Bernal L. Clinical experience in

extended cardiac monitoring with the QTM

satellite wireless system. Rev Colomb

Cardiol.2018;25(3):e1725.doi:10.1016/j.rccar.

2017.09.00

AUTOMATED DETECTION OF COVID-19 INFECTEDLESION ON COMPUTED
TOMOGRAPHY IMAGES USING FASTER-RCNNS

Siti Nurmaini, Member, IAENG, Alexander Edo Tondas, Radiyati Umi Partan, Muhammad NaufalRachmatullah,
Annisa Darmawahyuni, Firdaus Firdaus, Bambang Tutuko, Rachmat Hidayat, Ade Iriani Sapitri

Abstract—The gold standard of a definitive test for the 2019 novel Corona Virus (SARS-CoV-2) is
reverse-transcription polymerase chain reaction (RT-PCR). However, its sensitivity ranged between
50% - 90% with high false negatives. Currently, false negatives are real clinical problems, caused by
the absence of antibodies formation during sampling(incubation period), impaired antibody formation in
immunocompromised patients, apart from sample acquirement technique and transportation issue.
Thus,repeated RT-PCR testing is often needed at the early stage of the disease, which may prove to
be difficult in a pandemic situation. In some research, the chest computed tomography (CT) image
was a rapid and reliable method to diagnose patients with suspected SARS-CoV-2 with higher
sensitivity compared to RT-PCR test, particularly the lab test is negative. In this study, 420 CT images
with 2,697 features from seven patients infected by SARS-CoV-2 and 200 CT images from healthy
individuals are used for analyzing. The convolutional neural networks (CNNs) with Faster-RCNNs
architecture isproposed to process the infected lesion detection. As a result, the proposed model shows
90.41% mAP, 99% accuracy, 98% sensitivity, 100% specificity, and 100% precision of classifier
performances. All performance produces a 100% score when it tests on external data CT image. It can
be seen from the detection result that Ground-glass opacities (GGO)-principallesions on CT images in
the peripheral and posterior sections of the lungs should be strongly suspected of developing SARS-
CoV-2 pneumonia. On average, it took less than 0.3 seconds per image to detect the abnormalities
from a CT image from data pre-processing to the output of the report. For a frontline clinical doctor, the
proposed model may be a promising, supplementary diagnostic process.

Index Terms—Lesion detection, SARS-CoV-2, COVID-19 pneumonia, CT images, Convolutional neural
networks, Faster-RCNNs

INTRODUCTION negatives need to be considered in diagnostic
NOVEL coronavirus, also known as SARS-CoV- interpretation due to unestablished validity in
2 or COVID-19 (Corona Virus Disease 2019) different laboratory measurements (variability in
has now infected more than 784,381 people in sensitivity and specificity). False-positive RT-
the world and spread to more than 190 PCR test results may result from the cross-
countries. These numbers were reported only reactive antibodies with other various other
in less than 100 days since it was first viruses (e.g. coronavirus, dengue virus), and
identified in Wuhan, China, at the end of past infection by a coronavirus. Meanwhile,
December 2019. SARS-CoV-2 belongs to the false-negative for COVID-19 can occur if the
genus Betacoronavirus, one of the genera of sampling test is carried out during the incubation
viruses in the Coronaviridae family, the type period when the antibodies are not yet formed,
that can infect the respiratory system [1]. In or in immunocompromised patients (impaired
many cases, this virus only causes mild antibody formation) [10][11].
respiratory infections, with flu-like symptoms
[2][3] However, the newly emerging virus can also Another alternative method that can diagnose
give rise to severe respiratory infections, such the SARS- CoV-2 is image analysis based on
as pneumonia, similar to the previously known the CT scans imaging data. The results of chest
Middle-East Respiratory Syndrome(MERS), and CT scans in SARS-CoV-2patients can show the
Severe Acute Respiratory Syndrome (SARS)[4]. severity of pneumonia from coronavirus. The
There are several techniques to detect the lungs are the organs most affected by the viral
novelCOVID-19 that infected in the lung, such reaction, withevidence of lesser damage in other
as reverse- transcription polymerase chain organs [6][12][13]. SARS-CoV-2causes exudative
reaction (RT-PCR) or gene sequencing for inflammation with pathological features similar to
respiratory or blood specimens, chest X- those caused by SARS and MERS syndrome.
Rays, and computed tomography (CT) chest However, the pulmonary fibrosis caused by the
images [2][3]. Currently, the gold standard of new coronavirus was not as serious as SARS
a definitive test for COVID-19 is the RT-PCR [14]. Blood vessels are damaged during the
test, which is considered to be highly specific. reaction of the human immune system against
Unfortunately, the sensitivity interval for the test viruses. It then allows fluid to leak into the lung
result only ranged between 60-97% [1][3][4]. tissue, which can be seen as white spots on CT
images in the chest. Some CT features of lung
Therefore, false negatives are a real clinical abnormality that can be observed during the
issue, and repeat testing may be appropriate in early phase of SARS-CoV-2 infection are as
one case to be sure about eliminating the follows : [2][4][6] (i) lung changes in ground-glass
disease. In some previous papers, CT scan opacities (GGO), consolidation, GGO, and
imaging had been shown to yield better reticular pattern, vacuolar sign, microvascular
sensitivity than RT-PCR testing [4][5]. A research dilation sign, fibrotic streaks, subpleural line, and
paper in the Radiology journal emphasized the subpleural transparent line; (ii) bronchial change
importance of CT to diagnose patients with in air bronchogram and bronchus distortion; and
suspected SARS-CoV-2 infection, particularly (iii) pleural change in thickening of pleura,
when the lab test result is negative [6][7]. pleural retraction sign, and pleural effusion.
Another study that used data from 1,014
patients who underwent both chest CT imaging However, the frequency of lung change,
and RT-PCR test within three days revealed especially of GGO scores, was significantly
sensitivities of 98% and 71%, respectively [8][9]. higher in early-phase disease than in advanced-
It implies that a large number of RT-PCR tests phase disease [11]. In the previous study, it was
cannot be identified quickly for treatment, determined, the majority of patients with SARS-
probably related to a difficulty in sampling CoV-2 pneumonia had a GGO turbidity rate of
acquirement techniques or transportation in a 86.1 %, or mixed and consolidated GGO of 64.4
pandemic situation. The false positives and false

%, and vascular enlargement in the lesion of it produces a robust model in shape, region, and
71.3 %. GGO are areas with interstitial spatial relation features. Specifically,
thickening in the lungs [6]. The lesions on CT Convolutional Neural Networks (CNNs) proven
images of SARS-CoV-2patients were more likely in automatic feature learning in medical
to have a peripheral distribution of 87.1%. In applications like endoscopy, cardiovascular,
addition, bilateral involvement, dominancy in cancer, lung infections and others [13]–[19]. It
lower lungs andmultifocality were 82.2%, 54.5% indicates the feature learning-based CNNs with
and 54.5%, respectively [12]. The architectural CT images in the chest for detection of the
distortion, traction bronchiectasis, and pleural SARS-CoV-2produce good performances [20]–[23].
effusion, possibly reflect the viral load, virulence The main features of SARS-CoV-2are the
or the level of pathogenicity of SARS-CoV-2 bilateral distribution of patchy shadows and
[12]. Such a condition can be evaluated from lung GGO [2][7]. According to hallmarks, the CNNs
images. Chest CT severity score (CT-SS) can can process the automated feature learning that
help to evaluate the severity and level of might be difficult for the conventional visual
pneumonia caused by the coronavirus [13]; recognition approach. While typical CT images-
however, there were variabilities in the help to screen suspected cases early on, the
performance of radiologists to differentiate small CT images for early screening pneumonia
SARS-CoV-2with viral pneumonia [14]. Currently, caused by infectious and inflammatory lung
deep learning (DL), one of the Artificial diseases are difficult to be detected [8][9][11].
Intelligence (AI) approaches, has been proposed Hence, the deep investigation to diagnosis the
as a potential technique in CT image region SARS-CoV-2pneumonia is desirable.
detection. Based on automatic feature learning,

MATERIAL AND METHOD
CT image patterns of viral pneumonia caused by SARS- CoV-2shown by different pathogens. The
radiologists differentiate viral infections to diagnose SARS-CoV-2 pneumonia by using CT imaging
findings of these emerging pathogens. In this study the framework to make the CTpattern interpretation
based on framework is presented in Fig. 1.

Fig. 1. The framework of Faster-RCNNs investigation for SARS-CoV-2 infected lesion detection

The CT image as raw data, and it is delineation CNNs-based region detection. When the region
by radiologists to assign the region infected by is detected, it is validated again by radiologists
viral. All the delineation data is divided into data to ensure the result for making the robustness of
training, validation, and testing. The AI approach CNNs model.
uses for learning with deep structure by using

A. Data Preparation delineation for the infected region before the
A number of data used in this study were learning process. In this study, the CT patterns
collected from 419 CT scan images SARS-CoV- of viral pneumonia only are related to the
2 infected cases from seven patients, and 200 pathogenesis of SARS- CoV-2 infections by
CT scan images of normal condition from two using region of interest (ROI) images to define
patients from a well-known website [23]. The the inflammatory lesions based on ground-glass
view of CT images from 4 ways, axial lung opacity, mosaic sign, and interlobular septal
window, axial non-contrast, coronal lung thickening as presented in Fig. 2 (a), and health
window, and coronal non-contrast. In this study, condition in Fig. 2 (b).
three radiologist experts create manual

(a)

(b)
Fig. 2. (a) Sample of CT scan images SARS-CoV-2 infected; (b)sample of CT scan images of
healthy condition.

B. Pre-processing images to extract effective pulmonary regions;
In this stage, data delineation from the (ii) multiple candidate image cubes were
radiologist is annotated again for custom classes segmented using a 2D CNN model, the center
of CT images dataset with bounding boxes image was collected along with the two
process as presented in Fig. 3 (a). In addition, neighbors of each cube for further steps; (iii) an
an example of delineation by the software is image detection model was used to categorize
presented in Fig. 3 (b). The input data which all the image patches into two types normal and
contains a bunch of images with their bounding infected with average precision (AP) score to
boxes information. All data infected by SARS- select the confidence value of candidate
CoV-2 were about 419 images with a pixel size recognition as a whole; and (iv). The
that varies from 256 x 256 pixels until 1500 x probability function uses to measure the lesion
1800 pixels. Prior to the learning process, the as normal or infected by SARS-CoV-2. The
data is splitting into two, i.e. (1) 80 % for training ROI is sketched on the CT images based on at
about 335 images and (2) 20% remaining for least three features of pneumonia, including
testing about 84 images. All process to detect ground-glass opacity, mosaic sign, and
infected lesion by SARS-CoV-2 is presented in interlobular septal thickening [2][8][11]. For a ROI, it
Fig. 4. There are four stages of completing the issized approximately from 600x600 pixels up to
whole process, i.e. (i) pre-processing of the CT 1024x1024 pixel.

(a) Delineation by radiologist

(b) A delineation by software to create the data for training, and testing

Fig. 3. CT scan images preparation for the learning process.

Fig. 4. The process of infected lesion detection. input image, which is fed into the backbone
CNNs. The backbone network is used to
C. Faster-RCNNs recognize the infected region caused by SARS-
CNNs, with the strong ability of nonlinear CoV-2 in the lung as a pneumonia condition.
modeling, have extensive applications in VGG16 and ResNet50 architecture are created
medical image processing as well [24][25]. One of to produce a high accuracy of region detection.
the CNNs architecture is based on an object The Faster R-CNNs detector also consists of a
detection approach for exploring the infected backbone of the CNNs, a pooling layer of the
region by SARS-CoV-2, named Faster-RCNNs ROI and fully connected layers followed by two
used in this study [26]. The proposed sibling branches for classification and bounding
architecture with Faster R-CNNs is presented in box regression. A detail of the Faster R-CNNs
Fig. 5. The architecture consists of the region general architecture, as shown in Fig. 5.
proposal network (RPN) as an algorithm for
region proposal and the Fast R-CNNs as a
network for detectors. The RPN starts with the

Fig. 5. The Faster-RCNNs architecture.

The proposed model must tune the select the best model. For the RPNs backbone
hyperparameter toreduce the false negative and modeling with hyperparameters, 50 epochs to
false positive of bounding box detection. 200 epochs, one batch size, 0.0001 learning
Therefore this system is warranted to be further rate, and 0.9 momentum. For achieving the best
optimized and tested. Training and testing both sensitivity and specificity of the CNNs detection,
region proposal and object detection networks Stochastic Gradient Descent (SGD), and Adam
on CT images are developed of a single scale of optimizer utilized for the pre-trained VGG16. All
about 600 pixels and up to 1024 pixels. Two hyperparameter tuning for feature learning of
backbone CNNs architecture models, such as 2,697 “bounding box” from 275 images. In the
VGG16 and ResNet50 will be compared to fine-tuning for Faster RCNN architecture

TABLE I
FASTER-RCNNS STRUCTURE

Layer Number of Filter Kernel size Stride

Convolution 64 3x3 1

Convolution 64 3x3 1
Max Polling - 2x2 2
Convolution 128 3x3 1
Convolution 128 3x3 1
Max Polling - 2x2 2
Convolution 256 3x3 1
Convolution 256 3x3 1
Convolution 256 3x3 1
Max Polling - 2x2 2
Convolution 512 3x3 1
Convolution 512 3x3 1
Convolution 512 3x3 1
Max Polling - 2x2 2
Convolution 512 3x3 1
Convolution 512 3x3 1
Convolution 512 3x3 1

was used anchor box scale: 128, 256, and 512 pixels with ratio: (1:1), (1:2), and (2:1) respectively. All
bounding boxis the same object as lung region infected by SARS-CoV-2, therefore the overlap RPNs

is set at min = 0.3 and max =0.5, also overlap classifier ROI is set a min = 0.1 and max = 0.5. The
Faster-RCNNs structure, as seen in Table 1. In the last filter, 512 pixels used, induced by the large
stride, provide good results, although accuracy can be further improved with a smaller stride.

D. Performances Evaluation Metric the bounding box are associated as areas
In this study, the CNNs-based object detection infected by SARS-CoV-2 in different sizes. From
is based on regression and classification very small to large in one lung area. So, if the
process with two performances evaluation bounding box prediction cuts at least 0.3 part of
values such as, mean intersection over union the ground truth, it is assumed that the infected
(mIoU), and mean average precision (mAP). area has been detected; therefore, the IoU
Theconcept of mIoU and mAP use to computes threshold is tuning between 0.3 to 0.5. The
how much the predicted boundary (bounding proposed model also calculates the accuracy
box) overlaps with the ground truth in the two (Acc.),sensitivity (Sens.), specificity (Spec.), and
process, good performance model that implies precision (Prec.) to know the whole object
two bounding boxes perfectly overlap. If the detection performances. To ensure the process
overlap value above the threshold are of object detection run in a good performance,
considered positive, and below the threshold are an Intel i9-9900k CPU together with NVIDIA
considered negative. The confidence score for GPU RTX 2080ti 11GB was used as server for
each object identified by the model in the image testing. The processing time depended largely
needs to be considered as well. It expected on the number of convolution layers in one CT
to have a confidence score above a certain image set.
threshold. In our proposed model, all objects in

RESULTS AND DISCUSSION 2,697 features as ground truth for a prediction
The gold standard for the SARS-CoV-2 and two physicians for validation of the result.
diagnosis has nucleic acid identification based By using Faster-RCNN, it achieved an accuracy
on RT-PCR for the existence of specific gene of about 99% in the best model (see Table 2).
sequences. However, the high number of false Other performance to indicate the proposed
negatives also occurs due to many factors, such model produce good detection such as 98 %
as methodological drawbacks, disease stages, sensitivity, 100 % specificity, and 100%
and specimen collection methods that may delay precision for each image at the testingstage with
diagnosis and control of disease. Recent data 0.3 of IoU. Moreover, in this study, mAP is also
suggest that nucleic acidtesting accuracy is only used to measure the overlapping area between
around 50-90% [3][6][7]. Due to the limitation of ground truth and predicted image (see Table
RT-PCR test, there is an immediate need to look 3). In Fig. 3, the Faster-
for other simple alternative approaches which
can be used by frontline health care workers to
diagnose the disease rapidly and accurately.
Our study represents the AI-based region-CNNs
technologies for effectively screening of CT
images caused by SARS-CoV-2. In terms of CT
image lesion distribution, SARS-CoV-2 patients
were more likely to have peripheral distribution
(87.1%), bilateral involvement (82.2%), lower
lung predominance (54.5%), and multifocal
distribution (54.5%), consistent with the findings
of previous studies [10][11]. To develop a
detection algorithm, two physicians in
radiologists delineation the 420 images with

TABLE II

PRE-TRAINING AND FINE-TUNING PROCESS OF FASTER-RCNNS

ARCHITECTURE FOR SELECTING THE BEST MODEL WITH 200 EPOCHS

Training with Back-bone IoU Sens.(%) Spec(%) Prec (%) Acc. (%)

SARS-CoV-2 image VGG16 0.3 98 100 100 99
1024 pixels VGG16
SARS-CoV-2 image VGG16 0.4 98 85 96 96
1024 pixels
SARS-CoV-2 image VGG16 0.5 98 71 87 88
1024 pixels
SARS-CoV-2 and 0.3 95 90 98 94
healthy image 1024
pixels
SARS-CoV-2 and

healthy image 1024 VGG16 0.4 95 82 96 93
pixels VGG16 0.5 94 70 84 82
SARS-CoV-2 and
healthy image 1024

pixels

SARS-CoV-2 and VGG16 0.3 95 82 96 93
healthy image 600 VGG16 0.4 94 64 91 88
pixels VGG16 0.5 94 75 80 79
SARS-CoV-2 and ResNet50 0.3 90 86 98 90
healthy image 600 ResNet50 0.4 90 67 94 87
pixels ResNet50 0.5 88 62 76 72
SARS-CoV-2 and
healthy image 600
pixels
SARS-CoV-2 and
healthy image 1024
pixels
SARS-CoV-2 and
healthy image 1024
pixels
SARS-CoV-2 and
healthy image 1024
pixels

RCNN produce high mAP with VGG16 architecture as RPN backbone about 90.41% greater than
Resnet50 architecture backbone about 84.55%. It means VGG16 produce good

TABLE III
THE FASTER-RCNNS ARCHITECTURE PERFORMANCE FOR MEASURING THE OVERLAPPING BETWEEN GROUND

TRUTH ANDPREDICTED IMAGE IN MAP METRIC

Back bone IoU mAP (%)

VGG16 0.5 90.41
Resnet50 0.5 84.55

performance to detect the image infected by characteristics on CT images, including early
SARS-CoV-2. The result of CT images ground-glass and late-stage pulmonary
detection for two conditions, including SARS- consolidation opacities. The pathogens feature
CoV-2 infected and healthy, is presented in Fig. in CT imaging is associated with their specific
6. Multiple infected regions of SARS-CoV-2can pathogenesis of the lung lesion, such as patchy,
have single or double lesions, lower lobes are nodular, honeycomb, grid, or strips. Some of the
more commonly affected than upper and middle images in which the lesion density is often
lobes and the right middle lobe is the least irregular with the primary appearance of ground-
infectious (see Fig. 6 (a)). The result shows that glass opacity followed by interlobular or
the majority of COVID-19 cases have common intralobular septa thickening. The lesion can

also present aggregation and development of based region infected, all feature can detect
fiber stripes as paving stones. By using CNNs with some bounding box in the whole images.

Early stage

Medium stage

Late stage
(a) Infected condition in several early, medium, and late stage

(b) healthy condition

Fig. 6. The object detection in CT images results with two conditionsinfected lesion by SARS-CoV-2
and healthy.

In this study, the most surprising result, when the proposed model predicted the external data positive
and negative cases. The proposed model still managed to predict 100% for all metrics in terms of
accuracy, sensitivity, specificity, and precision certainty that the lungs with the infected region in the
images are positive for SARS-CoV-2as presented in Fig. 7

The amounts of studies have used the DL the results have been produced in this research
approach for the SARS-CoV-2 infected CT make it a promising good screening tool for
image lung classification given its superior SARS-CoV-2. During the existing outbreaks and
performance. However, such detection requires potential pandemics of COVID-19, the DL,
high sensitivity and specificity to avoid especially CNNs model, can potentially serve as
generating many false negatives that may a powerful tool for lung CT imaging screening.
cause excessive anxiety in the patient. The Using the supercomputer system, by using our
comparison between our model with another proposed model the time for detection, each
DL approach can be seen in Table 4. The result case only takes about 0.2-0.3 seconds, and can
found that our model improves performance. All be done remotely via a shared public platform.

Therefore further improvement of this method patients set is relatively small variation about
willgreatly shorten the diagnostic time for control nine patients, the predicted performances can
of disease. Hence, the proposed AI approach increase with increased training volume, (iii) the
should significantly contribute to the control of features of the CT images examined were in
COVID-19. The detection result can be reducing late-stage patients with severe lung lesions, (iv)
(i) the number of persons under investigations the research must enroll some cases of SARS-
for timely quarantine and treatment, and (ii) CoV-2 patients with early-stage for making the
human-to-human transmission. algorithm robustness, (v) this study not include
the data patient with a false negative in the RT-
However, there are some drawbacks due to the PCR test as a case in the robustness test, and
comparatively large number of variable objects (vi) comprehensive investigation with other virus
around 2,697, such as CT images reflect a infections can enable us to distinguish between
difficult task of classification, especially those SARS-CoV- 2 pneumonia and other lung
outside the lungs that are irrelevant to infections.
pneumonia diagnosis, (ii) the training data

CONCLUSION diagnostic tool for clinical doctors on the
To identify COVID-19 with higher sensitivity, the frontline.
combination of clinical symptoms, exposure
history, typical CT lung imaging features, and ACKNOWLEDGMENT
dynamic changes must be considered. Since We thank all our colleagues, especially all expert
single RT-PCR testing only produces a low physicians, who contributed significantly during
sensitivity, it should not exclude the diagnosis of the current study. We much appreciate the
SARS-CoV-2, especially if clinical suspicion is support of all our students in the Intelligent
high. To provide early recommendations based Systems Research Group (ISysRG), Universitas
on CT lung imaging, this study developed the Sriwijaya.
CNNs model with an object detection approach.
As depicted in the results, the proposed model REFERENCES
can detect the infected region by SARS-CoV- [1] V. M. Corman et al., “Detection of 2019
2 onCT chest radiography 99% accuracy, 98%
sensitivity, 100% specificity, and 100% novel coronavirus (2019- nCoV) by real-
precision. The testing by using external data time RT-PCR,” Eurosurveillance, vol. 25,
produce 100% accuracy, 100% sensitivity, and no. 3, 2020.
100% specificity. In the present study, the [2] Y. Yang et al., “Epidemiological and clinical
number of model samples was limited. features of the 2019 novel coronavirus
outbreak in China,” medRxiv, 2020.
Therefore, to achieved a better model with high- [3] D. K. W. Chu et al., “Molecular diagnosis of
performance, high efforts should be made to a novel coronavirus (2019-nCoV) causing
improve the bounding box, segmentation, and an outbreak of pneumonia,” Clin. Chem.,
classification model. Moreover, the efficiency of vol. 66, no. 4, pp. 549–555, 2020.
this generalization algorithm should be validated [4] N. Zhang et al., “Recent advances in the
with a larger set of data. Hence, the training and detection of respiratory virus infection in
testing of the number of samples should be humans,” J. Med. Virol., vol. 92, no. 4, pp.
expanded to improve performance in the future. 408– 417, 2020.
It can be concluded that our proposed CNNs [5] D. Wang et al., “Clinical characteristics of
architecture model can overcome the limited 138 hospitalized patients with 2019 novel
number of CT images data with good value in all coronavirus--infected pneumonia in Wuhan,
performance metrics in the object detection China,”Jama, 2020.
approach. It could be a helpful, additional [6] N. Chen et al., “Epidemiological and clinical
characteristics of 99 cases of 2019 novel

coronavirus pneumonia in Wuhan, China: Miernik, and A. Reiterer, “Application of
artificial neural networks for automated
a descriptive study,” Lancet, vol. 395, no. analysis of cystoscopic images: a review of
the current status and future prospects,”
10223, pp. 507–513, 2020. World J. Urol., pp. 1–10, 2020.
[18] P. Wang et al., “Development and
[7] Y. Fang et al., “Sensitivity of chest CT for validation of a deep-learning algorithm for
the detection of polyps during
COVID-19: comparison to RT-PCR,” colonoscopy,” Nat. Biomed. Eng., vol. 2, no.
10, pp. 741–748, 2018.
Radiology, p. 200432, 2020. [19] F. Shan+ et al., “Lung Infection
Quantification of COVID-19 in CT Images
[8] T. Ai et al., “Correlation of chest CT and with Deep Learning,” arXiv Prepr.
arXiv2003.04655, 2020.
RT-PCR testing in coronavirus disease [20] O. Gozes et al., “Rapid ai development
cycle for the coronavirus (covid-19)
2019 (COVID-19) in China: a report of 1014 pandemic: Initial results for automated
detection & patient monitoring using deep
cases,” Radiology, p. 200642, 2020. learning ct image analysis,” arXiv Prepr.
arXiv2003.05037, 2020.
[9] Q. Li et al., “Early transmission dynamics in [21] S. Wang et al., “A deep learning algorithm
using CT images to screen for Corona Virus
Wuhan, China, of novel coronavirus-- Disease (COVID-19),” medRxiv, 2020.
[22] H. J. Koo, S. Lim, J. Choe, S.-H. Choi, H.
infected pneumonia,” N. Engl. J. Med., Sung, and K.-H. Do, “Radiographic and CT
features of viral pneumonia,”
2020. Radiographics, vol. 38, no. 3, pp. 719–739,
2018.
[10] C. Huang et al., “Clinical features of [23] Radiopaedia.org, “COVID-19 Pneumonia.”
[24] Q. Zheng, M. Yang, X. Tian, X. Wang, and
patients infected with 2019 novel D. Wang, “Rethinking the Role of Activation
Functions in Deep Convolutional Neural
coronavirus in Wuhan, China,” Lancet, vol. Networks for Image Classification.,” Eng.
Lett., vol. 28, no. 1, 2020.
395, no. 10223,pp. 497–506, 2020. [25] M. Mohamadian, H. Afarideh, and F.
Babapour, “New 2d matrix- based neural
[11] W. Zhao, Z. Zhong, X. Xie, Q. Yu, and J. network for image processing applications,”
IAENG Int. J. Comput. Sci., vol. 42, no. 3,
Liu, “Relation between chest CT findings pp. 265–274, 2015.
[26] X. Xu et al., “Deep Learning System to
and clinical conditions of coronavirus Screen Coronavirus Disease 2019
Pneumonia,” arXiv Prepr. arXiv2002.09334,
disease (COVID-19) pneumonia: a 2020.
[27] Y. Song et al., “Deep learning Enables
multicenter study,” Am. J. Roentgenol., pp. Accurate Diagnosis of Novel Coronavirus
(COVID-19) with CT images,” medRxiv,
1–6, 2020. 2020.
[28] J. Chen et al., “Deep learning-based model
[12] R. Yang et al., “Chest CT Severity Score: for detecting 2019 novel coronavirus
pneumonia on high-resolution computed
An Imaging Tool for Assessing Severe

COVID-19,” Radiol. Cardiothorac. Imaging,

vol. 2, no. 2, p. e200047, 2020.

[13] H. X. Bai et al., “Performance of

radiologists in differentiating COVID-19

from viral pneumonia on chest CT,”

Radiology, p. 200823, 2020.

[14] P. Gómez, M. Semmler, A.

Schützenberger, C. Bohr, and M. Döllinger,

“Low-light image enhancement of high-

speed endoscopic videos using a

convolutional neural network,” Med. Biol.

Eng. Comput., vol. 57, no. 7, pp. 1451–

1463, 2019.

[15] J. Choe et al., “Deep Learning--based

Image Conversion of CT Reconstruction

Kernels Improves Radiomics

Reproducibility for Pulmonary Nodules or

Masses,” Radiology, vol. 292, no. 2, pp.

365–373, 2019.

[16] D. S. Kermany et al., “Identifying medical

diagnoses and treatable diseases by

image-based deep learning,” Cell, vol. 172,

no. 5, pp. 1122–1131, 2018.

[17] M. Negassi, R. Suarez-Ibarrola, S. Hein, A.

tomography: a prospective study,” Muhammad Naufal Rachmatullah is currently a
medRxiv, 2020. research assistant of Intelligent System
[29] A. Narin, C. Kaya, and Z. Pamuk, Research Group, Faculty of Computer Science,
“Automatic Detection of Coronavirus Universitas Sriwijaya, Indonesia. His research
Disease (COVID-19) Using X-ray Images interest includes Medical Imaging, Biomedical
and Deep Convolutional Neural Networks,” Signal and Engineering, Deep Learning, and
arXiv Prepr. arXiv2003.10849, 2020. Machine Learning.
[30] E. E.-D. Hemdan, M. A. Shouman, and M.
E. Karar, “COVIDX-Net: A Framework of Annisa Darmawahyuni is currently a research
Deep Learning Classifiers to Diagnose assistant of Intelligent System Research Group,
COVID-19 in X-Ray Images,” arXiv Prepr. Faculty of Computer Science, Universitas
arXiv2003.11055, 2020. Sriwijaya, Indonesia. Her research interest
[31] C. Zheng et al., “Deep Learning-based includes Biomedical Signal and Engineering,
Detection for COVID-19 from Chest CT Deep Learning, and Machine Learning.
using Weak Label,” medRxiv, 2020.
Firdaus Firdaus is currently is a lecturer and
Siti Nurmaini is currently a professor in the researcher in Intelligent System Research
Faculty of Computer Science, Universitas Group, Faculty of Computer Science,
Sriwijaya and IAENG member. Her research Universitas Sriwijaya, Indonesia. His research
interest, including Biomedical Engineering, Deep interest includes Text processing, Deep
Learning, Machine Learning, Image Processing, Learning, and Machine Learning.
Control systems, and Robotic.
Bambang Tutuko is currently is a lecturer and
Alexander Edo Tondas is currently a researcher in Intelligent System Research
cardiovascular specialist in Department of Group, Faculty of Computer Science,
Cardiology and Vascular Medicine, Mohammad Universitas Sriwijaya, Indonesia. His research
Hoesin General Hospital and Department of interest includes Robotics, Deep Learning, and
Biomedicine, Faculty of Medicine, Universitas Machine Learning.
Sriwijaya, Palembang, Indonesia. His research
research interest is Medicine, and Biomedical Rachmat Hidayat is currently lectrurer in
Signal and Engineering. Faculty of Medicine, Universitas Sriwijaya
Indonesia.
Radiyati Umi Partan is currently an internist at
at Muhammad Hoesin General Hospital, Ade Iriani Sapitri is currently a postgraduate
Indonesia. She is a lectrurer in Faculty of student of Faculty of Computer Science,
Medicine, Universitas Sriwijaya Indonesia. Her Universitas Sriwijaya, Indonesia. Her research
research interest is Medicine, and interest includes Medical Imaging, Deep
Pharmacology, Toxicology and Pharmaceutics. Learning, and Machine Learning.

ARRHYTHMIA RISK PROFILE AND VENTRICULAR REPOLARIZATION INDICES IN
COVID-19 PATIENTS: A SYSTEMATIC REVIEW AND META-ANALYSIS

Alexander Edo Tondas1,2, Rido Mulawarman3, Monica Trifitriana3 , Siti Nurmaini4, Irfannuddin3

1 Department of Cardiology and Vascular Medicine, Mohammad Hoesin General Hospital, Palembang, SumateraSelatan, Indonesia
2 Biomedicine Doctoral Program, Faculty of Medicine, Universitas Sriwijaya, Palembang, Indonesia
3 Faculty of Medicine, Universitas Sriwijaya Palembang, Indonesia
4 Intelligent System Research Group, Universitas Sriwijaya, Palembang, Indonesia

ABSTRACT
Introduction: Coronavirus disease 2019 (COVID-19) has been associated with cardiac arrhythmias.
Several electrocardiographic markers have been used to predict the risk of arrhythmia in patients with
COVID-19. We aim to investigate the electrocardiographic (ECG) ventricular repolarization indices in
patients with COVID-19.
Methodology: We performed a comprehensive systematic literature search from PubMed, EuropePMC,
SCOPUS, Cochrane Central Database,and Google Scholar Preprint Servers. The primary endpoints of this
search were: Tp-e (T-peak-to-T-end) interval, QTd (QT dispersion), and Tp-e/QTc ratio in patients with
newly diagnosed COVID-19 from inception up until August 2020.
Results: There were a total of 241 patients from 2 studies. Meta-analysis showed that Tp-e/QTc ratio was
higher in COVID-19 group (mean difference 0.02 [0.01, 0.02], p < 0.001; I2: 18%,). Tp-e interval was more
prolonged in COVID-19 group (mean difference 7.76 [3.11, 12.41],p < 0.001; I2: 80%) compared to control
group. QT dispersion (QTd) also was increased in COVID-19 group (mean difference 1.22 [0.61, 1.83], p <
0.001 ; I2:30%).
Conclusions: Several electrocardiographic markers including Tp-e/QTc, Tp-e interval, and QTd are
significantly increased in patients with COVID-19.

Key words: COVID-19; SARS-CoV-2; arrhythmia; ventricular repolarization; ECG; meta-analysis; T-peak-
to-T-end.
J Infect Dev Ctries 2021; 15(2):224-229. doi:10.3855/jidc.13922

(Received 16 September 2020 – Accepted 30 November 2020)

Introduction article was independently screened by three
The COVID-19 pandemic has spread worldwide, authors (AET, RM, and MT). After eliminating
affecting 21.2 million and taken a death toll of any irrelevant articles, the full texts were then
761,000 people, by the time this paper was thoroughly assessed according to the criteria for
written [1]. Althoughthe new virus (SARS-CoV-2) inclusion or exclusion below. The search was
is mostly associated with respiratory symptoms, finalized on 15 August 2020. This systematic
a recent paper has highlighted the role of review and meta-analysis are compliant with
cardiac injury in mortality and critically ill Preferred Reporting Items for Systematic
pneumonia in COVID-19 patients [2]. The Reviews and Meta- Analyses (PRISMA). The
pathophysiology of COVID-19 myocarditis study process can beappreciated in Figure 1.
probably roots from direct viral injury and cardiac
injury due to the host’s immune response, which Inclusion and exclusion criteria
is the cytokinestorm [3,4]. Previously, ventricular All research articles that described adult patients
arrhythmias have been reported to be quite diagnosed with COVID-19, together with
frequent in viral myocarditis or pericarditis [5], information on arrhythmogenic, repolarization,
and a significant rise of its incidencewas noted in and ECG parameters were included in this
patients with implantable cardioverter- study. We excluded articles other than original
defibrillators during the influenza epidemics [6]. research, case series with samples below 20,
case reports, articles on research in pediatric
Arrhythmias were observed in 19% of COVID-19 populations (age ≤17 years), and non-English
patients, according to a recent meta-analysis, language articles.
and their presence was associated with a poorer
outcome [7]. Some novel electrocardiography Data extraction
(ECG) markers, such as Tpeak-to-Tend (Tp-e), Data were extracted separately by three authors
QT dispersion (QTd), and Tp- e/QT ratio have (AET, RM, and MT) using a standardized
been shown to reflect transmural ventricular method to identify the relevant study
dispersion of repolarization or the repolarization characteristics and the outcome of interest.
heterogeneity and defined as predictors of risk Study characteristics included the author, year,
for ventricular arrhythmias and sudden cardiac study design, age, sex, cardiovasculardiseases,
death in various clinical settings, including hypertension, diabetes mellitus, LVEF, and
myocarditis [8–10]. This systematic review and smoking (former and current).
meta-analysis aimed to evaluate ventricular
repolarization parameters in treatment-naive Statistical analysis
COVID-19 patients compared to healthy To perform the meta-analysis, we used RevMan
individuals. version 5.3 software (Cochrane Collaboration).
We used the odds ratio (OR) and a 95% CI as a
Methodology pooled measure for dichotomous data. We used
Search strategy and study selection mean difference (MD) and its SD as a pooled
A systematic literature search was performed, measure for the continuous data. The
using PubMed, EuropePMC, SCOPUS, Google inconsistency index (I2) test which ranges from
Scholar Preprint Servers, and the Cochrane 0% to 100% was used to assess heterogeneity
Central Database with the search terms 1) across studies. A value above 50% or p <0.05
“COVID-19” OR “SARS- CoV-2” AND indicates statistically significant heterogeneity.
“arrhythmogenic”; 2) “COVID-19” OR “SARS-
CoV2” AND “Repolarization”; 3) “COVID- 19”
OR “SARS-CoV-2” AND “ECG”. After the
removal of duplicates, the abstract for each

We used the Mantel-Haenzsel method (for OR), and theInverse Variance method (for MD) with a fixed-effect
model for meta-analysis, and a random effect model was used in case of heterogeneity. All p values
were 2-tailed with a statistical significance set at 0.05 or below.

Figure 1. PRISMA flowchart

Results of 241 patients (Yenerçağ, 2020 and Öztürk,
Study selection 2020) [11,12].
We found a total of 280 records, of which 178
remained after the removal of duplicates; 152 Study Characteristics
records were excluded after screening the A total of 241 patients from two case-control
title/abstracts. After assessing 26 full texts for studies were included. Patients enrolled in these
eligibility, we excluded 24 for the following studieshave similar gender characteristics with a
reasons: 1) no data on arrhythmia; 2) literature mean age ranging from 48-56 years old, as seen
review only; 3) no data on ECG parameters. We in Table 1. The control groups from all studies
included 2 studies in the qualitative synthesis were healthy individuals (not COVID-19), age-
and in the meta-analysis, which included a total matched with the COVID-19 group.

Table 1. Characteristics of the included studies.

Author,year Study Design Sample (n) Age (years) Gender Hypertension(%) DM (%) Smoking (%) LVEF(%)
(male/female)
59.9 ± 2vs
Yenerçağ, cross-sectional, 75 vs 75 55.5 ± 17.1 vs 39/36 52 vs 54 36 37 60.9 ±
2020 single-centre 51 vs 40 50.2 ± 16.6 vs 41/34 (p = 0.885) vs 33 vs 40 2.1
studies (p = 0.0053) (p = 0.777) (p = 0.273) (p = 0.478)
Öztürk,2020 cross-sectional, 49.2 ± 16.7 vs 58.5 ± 5.4
double-blinded 47.9 ± 14.9 29/22 vs 26/14 11.7 19.6 Vs 60 ± 4.3
studies (p = 0.39) (p = 0.431) 11.8 vs 10.0(p = 0.789) vs 7.5 vs 12.5
(p = 0.384)
(p = 0.951)

Table 2. Assessment of publication bias using the Newcastle-Ottawa Scale (NOS).

Selection Comparability Exposure Total

Study Is the case Representativ Selection of Definition of Ascertainment of Same method of Non- Response *******
definition eness of the Controls Controls exposure ascertainment for rate ******
adequate? cases and controls
cases

Yenerçağ et * * * _ ** * *
al, 2020 *

Ozturk et al, * * _ _ ** *
2020

Figure 2. Tp-e/QTc ratio showed to be higher in COVID-19 patients group.

Figure 3. Mean Tp-e interval showed to be longer in COVID-19 patients group.

Figure 4. QTd was found to be longer in the COVID-19 patients' group.

The studies reported comorbidities such as (PCT) testing was used in both studies for
hypertension, diabetes mellitus, and smoking establishing the diagnosis of COVID-19.
history, which are proportionally similar in the
COVID-19 group and control group. The mean Tp-e/QTc Ratio and COVID-19
values of left ventricular ejection fraction (LVEF) Pooled analysis of the electrocardiographic
were similar in COVID-19 groups and control measurement showed increased Tp-e/QTc ratio
groups. All included studies reported baseline in COVID-19 group compared to control group
ventricular repolarization parameters relevant to with meandifference of 0.02 [95% CI 0.01-0.02],
arrhythmogenic risk: Tp-e/QTc ratio, Tp-e p < 0.001; I2:18%, p = 0.27 (Figure 2).
interval, and QTd. Polymerase chain reaction

Tp-e interval and Covid-19 myocardial cell layers: endocardial, epicardial,
Both studies from Yenerçağ et al and Öztürk et and mid-myocardial M cells, which have the
al found that the mean Tp-e interval to be longest APD, prone to further prolongation by
significantly longer in the COVID-19 group with a external factors. This phenomenon can be
mean difference of 7.76 milliseconds [95% CI extrapolated to the surface electrocardiogram
3.11-12.41], p < 0.001; I2: (ECG), where repolarization of the epicardial
80%, p = 0.03. (Figure 3) layer ends at peak of the T-wave but
repolarization of the M cells continues until the
QTd and Covid-19 end of T wave. Thus, measuring the time
Meta-analysis showed that prolongation of QTd between the peak and the end of the T
interval was associated with the COVID-19 wave reflects TDR, and the Tp-e interval has
group. Standardized mean difference was 1.22 been defined as a novel ECG marker for
milliseconds [95% CI 0.61-1.83], p < 0.001; arrhythmia and SCD vulnerability, beyond the
I2:30%, p = 0.23. (Figure 4). QT interval only [9,17]. The optimal method to
measure the Tp-e interval is to derive it from the
Publication Bias precordial ECG leads [18], which has been
We used the Newcastle-Ottawa Scale (NOS) for implemented in the two included studies.
case-control studies to assess publication bias
(Table 2). All of the included studies have 6-7 Tp-e and Tp-e/QT ratios have been used as
stars, indicating high-quality papers with a low event predictors in numerous clinical scenarios,
risk of publication bias. such as heart failure, Brugada syndrome,
hypertrophic cardiomyopathy, bradyarrhythmia,
Discussion and even in the general population [10,19–21].
With this meta-analysis, we can show In our study, both the Tp-e interval and Tp-e/QT
incremental changes in some ratio were significantly higher in the COVID-19
electrocardiographic parameters of ventricular group compared to the control. Even so, the
repolarization indices, which may become a values were still less than 100 milliseconds,
plausible explanation for the increased incidence which is the cut-off for higher risk in heart failure
of arrhythmias in previous reports [13–15]. The and myocardial infarction population [22].
systemic inflammatory response in SARS-CoV2
infection and the accompanying cytokine release Prolongation of QT interval is the classic ECG
by the host immune system, particularly marker for predisposition to the occurrence of
interleukin-6 (IL-6) may exert a direct torsades de pointes (TdP) or malignant
electrophysiological effect on the myocardium. arrhythmias and its monitoring can be simplified
IL-6 by itself directly inhibits the hERG-K+ for COVID-19 patients byusing handheld devices
channel, prolonging ventricular action potential [23], especially for those who received QT-
duration (APD), and together with IL-1 and prolonging drugs for SARS-CoV-2, such as
tumor necrosis factor (TNF)-α, it can modulate hydroxychloroquine or azithromycin [24].
cardiomyocyte K+ and Ca2+ ion channels, However, unlike Tp-e, the QT interval is
causing the so-called inflammatory cardiac dependent on the heart rate and still has to be
channelopathies [16]. corrected [9], thus variations exist in manual
measurements [25]. Tp-e interval and Tp- e/QT
Reentrant ventricular arrhythmias are the main ratio were higher in patients with acquired QT
contributors to sudden cardiac death (SCD) and prolongation suffering from torsade de pointes
mortality in susceptible patients. One of the compared to those who did not, therefore
proposed mechanisms for reentry is transmural providing additional predictive value for
dispersion of repolarization (TDR) between three identifying high-risk patients [26].

QT dispersion (QTd) is defined as the difference on the potential of using ventricular
between the longest and shortest QT on a repolarization indices for risk stratification and
standard 12- lead ECG [27–31] Subjects with adding contemporary modalities to previously
QTc dispersion > 60 ms had a twofold risk for known subclinical severity assessment in
cardiac death or sudden death and a 40% COVID-19 patients [36,37].
increased mortality risk when compared to
those subjects with a QTc dispersion < 30 ms Conclusions
[32]. The normal values of QTd in the general Arrhythmogenic risk is higher in confirmed
population is controversial [33], explaining the COVID-19 patients, compared to the healthy
wide gap of baseline values of QTd in the two population, as reflected by the incremental
studies in this meta-analysis. The corrected Tp-e change in electrocardiographic markers such as
interval and Tp-e/QT ratio were found to be Tp-e interval, QTd, and Tp-e/QT ratio.
more accurate measurements of TDR compared
to the QT, QTd, and Tp-e intervals in patientswith Acknowledgements
chronic inflammatory fever [34]. This manuscript is a contribution dedicated to all
COVID-19 healthcare workers in Mohammad
Limitations Hoesin General Hospital, Palembang, Indonesia,
This systematic review and meta-analysis have who shared their spirit and inspiration in fighting
several limitations. First, the studies were mostly this pandemic.
case- control and did not show a direct
correlation of repolarization indices with the References
incidence of arrhythmias afterward. This is, 1. WHO (2020) Novel Coronavirus (2019-nCoV)
however, quite a challenge, since the COVID-19
pandemic has created a reluctance to monitor Situation Report - 1. WHO Bull
the patient for a longer period, for instance using 2. Santoso A, Pranata R, Wibowo A, Al-Farabi
ambulatory ECG (AECGs) or Holter monitors,
therefore the true occurrence of arrhythmias MJ, Huang I, Antariksa B (2020) Cardiac
may be missed in most studies on this subject. injury is associated with mortalityand critically
Second, although the risk of publication bias is ill pneumonia in COVID-19: A meta-analysis.
low as shown by the Newcastle-Ottawa Scale, Am J Emerg Med S0735-6757(20) 30280-1.
the studies are quiteheterogenous, especially for 3. Siripanthong B, Nazarian S, Muser D, Deo R,
the analysis of mean Tp-e interval (I2>50%, p < Santangeli P, Khanji MY, Cooper LT, Chahal
0.05). Probably this is due to the small number CAA (2020) Recognizing COVID-19–related
of studies included in this manuscript. Third, the myocarditis: The possible pathophysiology
studies did not report electrolyte levels, which and proposed guideline for diagnosis and
can be a confounding factor for repolarization management. Hear Rhythm 17: 1463-1471.
changes. Fourth, the population of studies are 4. Lazzerini PE, Boutjdir M, Capecchi PL (2020)
limited toTurkish, and may not be generalizable COVID-19, Arrhythmic Risk, and
to other populations. Lastly, this study has not Inflammation: Mind the Gap! Circulation 7–9.
answered the arrhythmogenic risk related to 5. Baksi AJ, Kanaganayagam GS, Prasad SK
atrial arrhythmias, including atrial fibrillation, (2015) Arrhythmias in Viral Myocarditis and
which comprised more than 90% of all Pericarditis. Card Electrophysiol Clin 7: 269-
arrhythmias documented in COVID-19 patients 281.
[35]. Future studies should also be directed to 6. Madjid M, Connolly AT, Nabutovsky Y,
assess atrial repolarization indices. Safavi-Naeini P, Razavi M, Miller CC (2019)
Nevertheless, our study provided new insights Effect of High Influenza Activity on Risk of
Ventricular Arrhythmias Requiring Therapy in
Patients With Implantable Cardiac
Defibrillators and Cardiac Resynchronization

Therapy Defibrillators. Am J Cardiol. Jul 16.Lazzerini PE, Laghi-Pasini F, Boutjdir M,
Capecchi PL (2019) Cardioimmunology of
1;124(1):44-50. arrhythmias: the role of autoimmune and
inflammatory cardiac channelopathies. Nat
7. Pranata R, Huang I, Raharjo SB (2020) Rev Immunol 19: 63-64.

Incidence and impact of cardiac arrhythmias 17.Son O, Boduroglu Y (2019) Comparing of Tp-
Te interval andTp-Te/Qt ratio in patients with
in coronavirus disease 2019 (COVID- 19): A preserved, mid-range and reduced ejection
fraction heart failure. Open Access Maced J
systematic review and meta-analysis. Indian Med Sci 7: 752-759.

Pacing Electrophysiol J 2019: 1–6 . 18.Rosenthal TM, Masvidal D, Abi Samra FM,
Bernard ML, Khatib S, Polin GM, Rogers PA,
8. Tse G, Yan BP (2017) Traditional and novel Xue JQ, Morin DP (2018) Optimal method of
measuring the T-peak to T-end interval for
electrocardiographic conduction and risk stratification in primary prevention.
Europace 20: 698– 705.
repolarization markers of sudden cardiac
19.Panikkath R, Reinier K, Uy-Evanado A,
death. Europace 19: 712–721. Teodorescu C, Hattenhauer J, Mariani R,
Gunson K, Jui J, Chugh SS (2011)
9. Prenner SB, Shah SJ, Goldberger JJ, Sauer Prolonged tpeak-to-tend interval on the
resting ECG is associated with increased risk
AJ (2016) Repolarization Heterogeneity: of sudden cardiac death. Circ Arrhythmia
Electrophysiol 4: 441–447.
Beyond the QT Interval. J Am Heart Assoc 5:
20.Abdelghani SA, Rosenthal TM, Morin DP
e003607. (2016) Surface electrocardiogram predictors
of sudden cardiac arrest. Ochsner J 16: 280–
10.Ucar FM, Ozturk C, Yllmaztepe MA (2019) 289.

Evaluation of Tp-e interval, Tp-e/QT ratio and 21.Tähtisalo HS, Hiltunen TP, Kenttä T, Junttila
J, Oikarinen L, Virolainen J, Kontula KK,
Tp-e/QTc ratio in patients with acute Porthan K (2020) Effect of four classes of
antihypertensive drugs on cardiac
myocarditis. BMC Cardiovasc Disord 19:232. repolarization heterogeneity: A double-blind
rotational study. PLoS One 15: 1–13.
11.Yenerçağ M, Arslan U, Doğduş M, Günal Ö,
22.Morin DP, Saad MN, Shams OF, Owen JS,
Öztürk ÇE, Aksan G, Erdoğan G, Gül S, Xue JQ, Abi-Samra FM, Khatib S, Nelson-
Twakor OS, Milani R V (2012) Relationships
Yontar OC, Şen A (2020) Evaluation of between the T-peak to T-end interval,
ventricular tachyarrhythmia, and death in left
electrocardiographic ventricular ventricular systolic dysfunction. Europace 14:
1172-1179.
repolarization variables in patients with newly
23.Tondas AE, Halim RA, Guyanto M (2020)
diagnosedCOVID-19.J Electrocardiol 62:5–9. Minimal or No Touch Electrocardiography
Recording and Remote Heart Rhythm
12.Öztürk F, Karaduman M, Çoldur R, İncecik Ş, Monitoring during COVID-19 Pandemic Era.
Indones J Cardiol 41: 133-41. doi:
Güneş Y, Tuncer M (2020) Interpretation of 10.30701/ijc.1010.

arrhythmogenic effects of COVID-19 disease

through ECG. Aging Male 1-4.

13.Singh S, Desai R (2020) COVID-19 and new-

onset arrhythmia.J Arrhythm 36: 547-548.

14.Kochav SM, Coromilas E, Nalbandian A,

Ranard LS, Gupta A, Chung MK,

Gopinathannair R, Biviano AB, Garan H, Wan

EY (2020) Cardiac Arrhythmias in COVID-19

Infection. Circ Arrhythm Electrophysiol 13:

e008719.

15.Bhatla A, Mayer MM, Adusumalli S, Hyman

MC, Oh E, Tierney A, Moss J, Chahal AA,

Anesi G, Denduluri S, Domenico CM, Arkles

J, Abella BS, Bullinga JR, Callans DJ,Dixit S,

Epstein AE, Frankel DS, Garcia FC,

Kumareswaram R, Nazarian S, Riley MP,

Santangeli P, Schaller RD, Supple GE, Lin D,

Marchlinski F, Deo R (2020) COVID-19 and

cardiac arrhythmias. Hear Rhythm 17: 1439-

1444.

24.Purwowiyoto SL, Hermanto DY, Iqbal M patients with familial Mediterranean fever.
(2020) Managing QT prolongation in the Era Cardiovasc J Afr 28: 154- 158.
of Coronavirus Disease 2019 (COVID-19). 35.Gopinathannair R, Merchant FM, Lakkireddy
Indones J Cardiol. doi: 10.30701/ijc.1002. DR, Etheridge SP, Feigofsky S, Han JK,
Kabra R, Natale A, Poe S, Saha SA, Russo
25.Fosser C, Duczynski G, Agin M, Wicker P, AM (2020) COVID-19 and cardiac
Darpo B (2009) Comparison of manual and arrhythmias: a global perspective on
automated measurements of the QT interval arrhythmia characteristics and management
in healthy volunteers: An analysis of five strategies. J Interv Card Electrophysiol 59:
thorough QT studies. Clin Pharmacol Ther 329- 336.
86:503–506 . doi:10.1038/clpt.2009.34. 36.Nurmaini S, Tondas AE, Partan RU, Naufal M,
Darmawahyuni A, Firdaus F, Tutuko B,
26.Tse G, Gong M, Meng L, Wong CW, Hidayat R (2020) Automated Detection of
Bazoukis G, Chan MTV, Wong MCS, Letsas COVID-19 Infected Lesion on Computed
KP, Baranchuk A, Yan GX, Liu T, Wu WKK Tomography Images Using Faster-RCNNs.
(2018) Predictive value of Tpeak - Tend Eng Lett 28: 1295-1301.
indices for adverse outcomes in acquired QT 37.Pranata R, Tondas AE, Huang I, Lim MA,
prolongation: A meta- analysis. Front Physiol Siswanto BB, Meyer M, Mitrovic V (2020)
9: doi: 10.3389/fphys.2018.01226. Potential role of telemedicine in solving ST-
segment elevation dilemmas in remote areas
27.Glancy JM, Garratt CJ, Woods KL, De Bono during the COVID-19 pandemic. Am J Emerg
Med S0735- 6757(20)30495.
DP (1995) Three‐ Lead Measurement of QTc
Dispersion. J Cardiovasc Electrophysiol 6: Corresponding author
987–992 . Alexander Edo Tondas MD
28.Malik M, Camm AJ (1997) Mystery of QT(c) Department of Cardiology & Vascular Medicine,
interval dispersion. Am J Cardiol 79: 785– Mohammad Hoesin General Hospital,
787. Palembang
29.Satu E. Na¨ ppi, Vesa K. Virtanen, Heikki Tel : +6281373011090 (A.E.T)
H.T. Saha JTM, Pasternak and AI (2000) QT E-mail: [email protected]
c dispersion increases during hemodialysis
with low-calcium dialysate. 57: 2117–2122. Conflict of interests: No conflict of interests is
30.Okin PM, Devereux RB, Howard B V., declared.
Fabsitz RR, Lee ET, Welty TK (2000)
Assessment of QT interval and QT dispersion
for prediction of all-cause and cardiovascular
mortality in American Indians: The strong
heart study. Circulation 101: 61–66.
31.Sahu P (2000) QT dispersion in medicine:
electrophysiological Holy Grail or fool’s gold?
Qjm 93: 425–431.
32.Bańcarz A, Kobusiak-Prokopowicz M (1996)
QT dispersion. Przegla̧ d Lek 53: 687–689.
33.Castro-Torres Y (2015) Ventricular
repolarization markers for predicting
malignant arrhythmias in clinical practice.
World J Clin Cases 3: 705.
34.Karaman K, Karayakali M, Erken E, Demirtaş
A, Öztürk M, Altunkaş F, Arisoy A, Turan OE,
Ceyhan K, Çelik A (2017) Assessment of
myocardial repolarisation parameters in



EFFECT OF CORONAVIRUS DISEASE-19 PANDEMIC TO THE VOLUME OFTOTAL
HIP AND KNEE ARTHROPLASTY SURGICAL SERVICE: EXPERIENCE FROM A
SINGLE TERTIARY ORTHOPEDIC HOSPITAL IN INDONESIA

Asep Santoso1,2*, Gilang Persada1, Iwan Budiwan Anwar1,2, Tangkas SMHS Sibarani1,2, Ismail
Mariyanto1,2, Pamudji Utomo1,2

1Department of Orthopaedic and Traumatology, Universitas Sebelas Maret, Surakarta, Indonesia;
2Division of Adult Reconstructive Surgery and Sport Injury, Prof. Dr. R. Soeharso Orthopaedic Hospital,
Surakarta, Indonesia

ABSTRACT
The difficult situation of the coronavirus disease (COVID)-19 pandemic may affect to hip and knee
arthroplasty service. Retrospective study was performed to patients who received elective total hip/total
knee arthroplasty (THA/ TKA) from January to September 2020 at Prof. Dr. R. Soeharso Orthopaedic
Hospital, Surakarta, Indonesia. There were a total of 64 THA and 227 TKA from January to September
of 2020. There was an extreme decrease in the number of TKA surgery during April, May, and June in
2020. The effect of coronavirus disease-19 pandemic to the decrease of arthroplasty service was
mainly in the first 3 months period.

INTRODUCTION MATERIALS AND METHODS
The COVID-19 pandemic has been affecting This was a retrospective descriptive study of
orthopedic services all around the world [1]. Almost patients who received elective total hip/total
all aspects of orthopedic field include emergency knee arthroplasty (THA/TKA) during the period
service, outpatient service, inpatient service, and of January to September 2020 (9 months) at
elective surgical service, have been significantly Prof. Dr. R. Soeharso Orthopaedic Hospital,
altered [2]. Recommendations to postpone elective Surakarta, Indonesia. All elective THA/TKA
surgical service have been announced by the during the period were included in the study. We
national orthopedic society to reduce the COVID-19 collect the demographic data, total number of
spreading [3]. This difficult situation of course also surgery, number of surgery at each month, and
affects the surgical hip and knee arthroplasty service patient’s origin. We also evaluate similar data
[4]. Although there has been an increasing number during the period of January to September 2019
of arthroplasty in Indonesia in the previous year [5], (non-pandemic period) for comparison. Further,
the COVID-19 pandemic may also affect to hip and we also evaluate trends data of THA/TKA
knee arthroplasty service in Indonesia in the recent service at our institution during the past 6 years
situation. Several authors propose some period (2014–2019).Pre-operative screening at
recommendations to optimize the safety of the outpatient clinic was performed on all patients
elective surgical services while limiting the who will receive THA/TKA surgery includes
spread of COVID-19 [6], [7]. history and clinical examination, laboratory
The recommendation manages all aspects of examination, chest radiograph, rapid COVID- 19
the service include: patients, staff, facilities, test, and polymerase chain reaction (PCR) swab
surgery, and post-operative management [6]. test. Internist performed the decision to proceed
Another author also recommends an algorithm with elective surgery at our institution. Patients with
for resuming elective surgical service in the confirmed COVID- 19 based on PCR swab were
pandemic situation [7]. With the use of strict referred to a special referral hospital of COVID-19
screening rules before surgery, elective surgical for further care. There were a total of four active
service may still be possible to perform. In this arthroplasty surgeons at the division of adult
study, we tried to evaluate the effect of the reconstructive surgery at our hospital: Three
COVID-19 pandemic to hip and knee senior arthroplasty surgeons and one junior
arthroplasty surgical service in our institution. arthroplasty surgeon. Among the three senior

arthroplasty surgeons, two are aged more than not differentiate between primary or revision
60 years old. Posterolateral approach was the arthroplasty surgery in the data presented as the
most commonly used approach for THA at our number of revision surgery was very small and
institution. Medial parapatellar approach was the predicted to have no special impact in this recent
most commonly used approach in TKA. We did study.

RESULTS
There were a total of 64 THA and 227 TKA from January to September of 2020. The number of TKA
was increased compared to the similar period at 2019 with 175 TKA surgeries. A similar number of THA
surgeries were performed in 2019 and 2020. No difference in mean of age and proportion of gender
between 2019 and 2020 was found [Table 1].

Table 1: Demographic data (January–September)

Parameters 2019 2020

Mean age 57,39 58,27

Male/female 74/165 78/213

There was an increasing number of THA/TKA surgery since the past 6 years, from 2014 to 2019 at our
institution (Figure 1).

600

401
500 331

400
230

300 218

200 107

100 111 147 135
39 2016
97 77 2018 2019
0 28 2015 2017
2014

THA TKA

Figure 1: The trend of the total number of total hip and knee arthroplasty by year (2014–2019)

TKA was still predominantly performed surgery compared to THA. In 2019, the peak number of THA-
TKA surgeries occurred in April. Conversely, there was an extreme decrease of the number of THA-
TKA surgery during April, May, and June in 2020. The number of THA-TKA surgeries regaining its
number from July to September 2020 (Figure 2).

60

50 50 51

40 39 38 41 24 22 32 30 27
30 34 16 22 25

26 27 7 July Agustus
20
May June
17
10

0
January February March April

September

Figure 2: Comparison of total number of total hip arthroplasty during 2019 and 2020 (January–
September)

There were only 13, 7, and 14 TKA during April, May, and June 2020. In comparison, the total number of

TKA reaches its peak atApril 2019 with 41 surgeries (Figure 3).

Total Knee Arthroplasty (TKA)

45 40
40 42

35 34

30 31 27 26
23
25 25 24

20 20 18 19

15 13 15 134
10 10

57

0
January February March April May June July AgustusSeptember

2019 2020

Figure 3: Comparison of the total number of TKA during 2019 and 2020 (January–September)

THA surgeries showed more fluctuated data in the number of surgeries both in 2019 and 2020.
However, during the period of April and May 2020, the number of THA surgery was very small (Figure
4).

Total Hip Arthroplasty (THA)
14

12 12
11 11
10 10

8 9 99 8
8
7
66 6

44
23
2

00
January February March April May June July Agustus September

2019 2020

Figure 4: Comparison of the total number of total hip arthroplasty during 2019 and 2020 (January–

September)

Patient’s origin evaluations showed that patients who came from outside the city but still in the

same province were the predominant patient’s origin both in 2019 and 2020 (Figures 5 and 6). This is

followed by patients who came from the outside province in early 2020 (January, February, and March).

However, none of the patients come from another province in May 2020. Patients who come from

outside of the province regain its number in July, August, and September 2020.

25

20 21 18
19 19
15
15 14 14
10 12 12
12 8 10

8 5 8 2 8 4 5 5
5 3 April May 6 3 2

4 March 2 July AgustusSeptember
3 June
0
January February

Local region Outside region (Same province) Outside region (Different Province)

Figure 5: Origin of patients from January to September 2019

35

30

29

25 12 18
24 15 14
7
20 21 7 65 3 87 9 8
19 April 6
340 June
15 May July Agustus September

10 11 12 11 13

5 65
0

January February March

Local region Outside region (Same province) Outside region (Different Province)

Figure 6: Origin of patients from January to September 2020

DISCUSSION the disease end-stage during the COVID-19
Since the first confirmed case of COVID-19 in pandemic. About 90% of patients still want to
Indonesia in early March 2020, all aspects of have surgery as soon as possible [14]. We
medical services in Indonesia have been found a similar situation in which the patient’s
affected [8]. It is ranged from primary care expectation of receiving THA/TKA surgery was
service until surgery service [9], [10], [11]. There still high at our institution. As the hospital
were several previous papers that reported the preparedness to face the pandemic situation
effect of the COVID-19 pandemic to THA/TKA was better compared to the first 3 months
surgical service from other countries [12], [13]. period, the number of THA-TKA surgeries was
To the best of our knowledge, this is the first regained its number starting from July until
study to report the effect of the COVID-19 September 2020. A complete patient screening
pandemic on THA/TKA surgical service in a system combined with the availability of level-3
single institution in Indonesia. PPE gives a perception of safety to the person
involved in elective surgical service. A recent
The results of this study showed that the number short-term study at a national referral hospital in
of THA-TKA surgeries was decreased during the Indonesia showed that elective orthopedic
early period of the pandemic, especially in the surgery might not be associated with increased
first 3 months (April, May, and June 2020). In cases of COVID-19 cases [9]. However, further
this period, the majority of elective surgeries study is needed to confirm its findings.
were canceled and rescheduled to prevent the
possibility of severe acute respiratory syndrome- The decision in resuming elective surgical
CoV-2 spreading. One of the senior arthroplasty service, especially THA/TKA, needs a focused
surgeons who aged >60 years old was also strategy that depends on local condition of the
advised to be out of duty during April, May, and hospital [15]. Our hospital is not a referral
June 2020 due to the previous history of health hospital for COVID-19 cases. Patients with
problems and categorized as high-risk personal. confirmed COVID-19 based on PCR swab will be
On the other hand, all aspects of the hospital referred to a referral hospital of COVID-19 for
medical service also in the phase of mitigation further care. Recommendations related to the
and preparation for thepossibility of facing the safety of resuming THA/TKA surgical service
COVID-19 were fulfilled. The availability of have been released by several organizations
personal protective equipment (PPE) was [16], [17]. Several authors reported that the
another difficult problem in the early phase of decrease of THA/TKA surgery volume resulted in
the pandemic in Indonesia. The price of PPE an economic burden to all involved stakeholders
was high and difficult to find in the early period includes patients, physicians, and
pandemic. hospitals/health-care organizations [13], [14],
[15]. The burden will be higher, especially for a
A study in the United States showed that special orthopedic hospital like in our institution,
patients with hip/knee arthritis suffering from the as orthopedic service is the only core of
pain and continue to struggle with pain due to service in the hospital. Resuming THA/TKA

service may give an opportunity for economic orthopaedics. J Clin Orthop Trauma.
recovery while still ensuring the safety of
patients and health-care providers [18], [19]. 2020;11(3):498-99.
The patient’s origin is another issue to discuss in
this study. We found that the patient’s origin was https://doi.org/10.1016/j.jcot.2020.03.015
predominantly from another city but inside the
same province. The situation was similar to the PMid:32405218
previous year 2019. The government of
Indonesia did not apply a total whole country 3. Budhiparama NC, Rhatomy S, Santoso A.
“lockdown” during the COVID-19 pandemic.
However, restriction of people mobilization and Editorial: The inaugural issue of the hip and
transportation has been applied through a more
local restriction program at each city or province, knee journal. Hip Knee J 2020;1(1):1-3.
namely “Pembatasan Sosial Berskala Besar
(PSBB)” [20]. The program started in April 2020 4. Jella TK, Samuel LT, Acuña AJ, Emara AK,
in several regions of Indonesia. This also
affected the number of patients who visit our Kamath AF. Rapid decline in online search
institution during the same period.
There were limitations to this study. This was a queries for hip and knee arthroplasties
descriptive retrospective study, which may have
its own weakness. The data were only obtained concurrent with the COVID-19 pandemic. J
from a single orthopedic hospital, where the local
situation can be different compared to another Arthroplasty. 2020;35(10):2813-9.
institution, such as in a general hospital. The
period of evaluation was also only 9 months and https://doi.org/10.1016/j.arth.2020.05.051
the local situation of the hospital can be
dynamically change with regard to the pandemic PMid:32534864
situations. However, we believe that this study
could give some insight into the early effect of 5. Ramadhani RR, Mulyadi D, Nur Alam AI,
COVID-19 pandemic to THA/TKA surgical
service in our institution. Dharmayuda CG, Rhatomy S, Rahmadian

CONCLUSIONS R, et al. Trends on primary TKA in
The COVID-19 pandemic has an effect on the
volume of THA/TKA service at our institution. The government tertiary-referral hospitals 2019.
effect was mainly in the first 3-month period.
Comprehensive patient screening and complete Hip Knee J 2019;1(1):50-6.
level-3 PPE should be available before proceed
to resume elective THA/TKA service. 6. Gilat R, Haunschild ED, Tauro T, Cole BJ.

REFERENCES Recommendation to optimize safety of
1. Faria G, Onubogu IK, Tadros BJ, Relwani J.
elective surgical care while limiting the spread
Change in practice due to COVID-19 Early
experiences of a United Kingdom district of COVID-19: Primum non nocere. Arthrosc
general hospital in trauma and orthopaedics.
J Orthop. 2020;22:288-90. Sports Med Rehabil. 2020;2(3):e177-83.
https://doi.org/10.1016/j.jor.2020.06.004
PMid:32565644 https://doi.org/10.1016/j.asmr.2020.04.008
2. Haleem A, Javaid M, Vaishya R, Vaish A.
Effects of COVID-19 pandemic in the field of PMid:32342047

7. Iyengar KP, Jain VK, Vaish A, Vaishya R,

Maini L, Lal H. Post COVID-19: Planning

strategies to resume orthopaedic surgery-

challenges and considerations. J Clin Orthop

Trauma. 2020;11(Suppl 3):S291-5.

https://doi.org/10.1016/j. jcot.2020.04.028

8. Setiawaty V, Kosasih H, Mardian Y, Ajis E,

Prasetyowati EB, Siswanto S, et al. The

identification of first COVID-19 cluster in

Indonesia. Am J Trop Med Hyg.

2020;103(6):2339-42. https://

doi.org/10.4269/ajtmh.20-0554

PMid:33069265

9. Rhatomy S, Prasetyo TE. Impact of

COVID-19 on primary care visits: Lesson

learnt from the early pandemic period. J

Community Empowerment Health

2020;3(2):57918. https://doi.

org/10.22146/jcoemph.57918

10. Kamal AF, Widodo W, Kuncoro MW,

Karda IW, Prabowo Y, Singh G, et al. Does

elective orthopaedic surgery in pandemic era

increase risk of developing COVID-19? A

combined analysis of retrospective and

prospective study at Cipto Mangunkusumo COVID-19 pandemic: Guidelines developed
by the international consensus group (ICM).
Hospital, Jakarta, Indonesia. Ann Med Surg J Bone Joint Surg Am. 2020;102(14):1205-
12. https://doi.org/10.2106/jbjs.20.00844
(Lond). 2020;60:87- 91. PMid:32675662
18. O’Connor CM, Anoushiravani AA,
https://doi.org/10.1016/j.amsu.2020.10.015 DiCaprio MR, Healy WL, Iorio R. Economic
recovery after the COVID-19 pandemic:
11. Rhatomy S, Phatama KY, Rasyid FA, Resuming elective orthopedic surgery and
total joint arthroplasty. J Arthroplasty.
Mustamsir E. Changes in orthopedic 2020;35(7S):S32-36.
https://doi.org/10.1016/j. arth.2020.04.038
services in two indonesian tertiary-referral PMid:32345566
19. Kort NP, Zagra L, Barrena EG, Tandogan
hospitals during the coronavirus-19 RN, Thaler M, Berstock JR, et al. Resuming
hip and knee arthroplasty after COVID-19:
pandemic. Open Access Maced J Med Sci. ethical implications for wellbeing, safety and
the economy. Hip Int. 2020;30(5):492-9.
2020;8(T1):538-41. https://doi.org/10.3889/ https://doi.org/10.1177/1120700020941232
PMid:32635761
oamjms.2020.5491 20. Ariawan I, Jusril H. COVID-19 in Indonesia:
Where are we? Acta Med Indones.
12. D’Apolito R, Faraldi M, Ottaiano I, Zagra 2020;52(3):193-5. PMid:33020329.

L. Disruption of arthroplasty practice in an

orthopedic center in northern Italy during the

coronavirus disease 2019 pandemic. J

Arthroplasty. 2020;35(7S):S6-9.

https://doi.org/10.1016/j.arth.2020.04.057

PMid:32370923

13. Bedard NA, Elkins JM, Brown TS. Effect

of COVID-19 on hip and knee arthroplasty

surgical volume in the United States. J

Arthroplasty. 2020;35(7S):S45-8.

https://doi.org/10.1016/j. arth.2020.04.060

PMid:32381441

14. Brown TS, Bedard NA, Rojas EO,

Anthony CA, Schwarzkopf R, Barnes CL, et

al. The effect of the COVID-19 pandemic on

electively scheduled hip and knee

arthroplasty patients in the United States. J

Arthroplasty. 2020;35(7S):S49-55. https://doi.

org/10.1016/j.arth.2020.04.052

PMid:32376163

15. Zeegen EN, Yates AJ, Jevsevar DS. After

the COVID-19 pandemic: Returning to

normalcy or returning to a new normal? J

Arthroplasty. 2020;35(7S):S37-41.

https://doi.org/10.1016/j. arth.2020.04.040

PMid:32376171

16. Kort NP, Barrena EG, Bédard M, Donell

S, Epinette JA, Gomberg B, et al. Resuming

elective hip and knee arthroplasty after the

first phase of the SARS-CoV-2 pandemic: The

European Hip Society and European Knee

Associates recommendations. Knee Surg

Sports Traumatol Arthrosc. 2020;28(9):2730-

46. https://doi.org/10.1007/s00167-020-

06233-9 PMid:32844246

17. Parvizi J, Gehrke T, Krueger CA, Chisari

E, Citak M, Van Onsem S, et al. Resuming

elective orthopaedic surgery during the



LAPORAN KASUS KOINFEKSI dan covid-19 sehingga pasien dapat
TUBERKULOSIS RESISTEN OBAT ditatalaksana dengan tepat dan dapat dicegah
outcome yang buruk.
DAN COVID-19
PENDAHULUAN
Astuti Setyawati, Hasto Nugroho, Ign Kejadian koinfeksi TB dan covid-19 masih
Widiyawati, Jatu Aviani jarang dilaporkan. Tuberkulosis resisten obat
dan covid-19 memiliki gejala dan tanda yang
SMF Paru, RS Paru dr Ario Wirawan hampir sama. Gejala meliputi demam, batuk,
Jalan Hasanudin no 806, Sidomukti, Salatiga, sesak, dan nyeri dada. Koinfeksi TB resisten
dan covid-19 dapat memperburuk outcome.
Jawa Tengah Pasienn TB RO mendapatkan pengobatan
Cp +085200691895, email: dengan jumlah yang banyak dan jangka lama,
[email protected]. sehingga efek samping yang ditimbulkan juga
lebih banyak dan berat. Pada covid-19 terjadi
ABSTRAK badai sitokin yang memperburuk kondisi pasien
TB RO. Permasalahan juga timbul ketika pasien
Latar Belakang: Covid-19 adalah penyakit sudah terjadi perbaikan klinis dan melakukan
infeksi yang disebabkan oeh virus SARS cov-2. isolasi mandiri dirumah, yaitu saat pemberian
Penyakit covid-19 dapat terjadi pada semua dan pemantauan minum obat TB RO. Oleh
orang, diantaranya pasien tuberkulosis resisten karena itu dukungan dari keluarga dan fasilitas
obat (TB RO). Gejala penyakit dan gambaran kesehatan sangat penting dalam kelancaran dan
radiologis yang tidak khas pada covid-19 mirip kedisplinan pasien TB RO dalam minum OAT.
pada penyakit Tuberkulosis.
LAPORAN KASUS
Laporan Kasus: Laki-laki 65 tahun dengan Laki-laki berusia 65 tahun dating ke UGD RS
keluhan batuk darah, sesak napas, kejang, dan dengan keluhan batuk darah, sesak napas,
sulit diajak bicara. Pasien memiliki peyakit kejang, dan sulit diajak bicara. Pasien memiliki
Diabetes Melitus (DM)dan TB RO dalam terapi peyakit Diabetes Melitus (DM)dan TB RO dalam
obat anti tubekulosis (OAT) resisten obat terapi obat anti tubekulosis (OAT) resisten obat
selama 5 hari. Kontak Pasien Covid disangkal. selama 5 hari. Kontak Pasien Covid disangkal.
Faktor resiko covid didapatkan keluarga yang Faktor resiko covid didapatkan keuarga yang
berkunjung dari luar kota, anak-anak bekerja berkunjung dari luar kota, anak-anak bekerja
sebagai sopir truk, penjaga toko, dan karyawan sebagai sopir truk, pejaga toko, dan karyawan
pabrik. Pemeriksaan fisik didapatkan ronki di pabrik. Pemeriksaan fisik didapatkan ronki di
kedua lapang paru. Pemeriksaan rontgen kedua lapang paru. Pemeriksaan rontgen
thoraks dibandingkan sebelumnya didapatkan thoraks serial didapatkan penambahan infiltrat di
penambahan infiltrat di apeks paru kiri dan apeks paru kiri dan perifer kanan dibandingkan
perifer kanan. Pemeriksaan laboratorium darah rontgen sebelumnya. Pemeriksaan laboratorium
didapatkan lekosit 15.890/uL, trombosit darah didapatkan lekosit 15.890/uL, trombosit
455.00/uL, limfosit 3,1% dan monosit 6,3%, D 455.00/uL, limfosit 3,1% dan monosit 6,3%, D
Dimer 2115ng/ml FEU, rapid antibodi Ig G Dimer 2115ng/ml FEU, rapid ntibodi Ig G reaktif.
reaktif. Hasil pemeriksaan test cepat molekuler Hasil pemeriksaan test cepat molekuler (TCM)
(TCM) sputum didapatkan Mtb detected sputum didapatkan Mtb detected medium,
medium, Rifampisin resisten detected. Pasien Rifampisin resisten detected. Hasi LPA
dirawat di ruang isolasi dan dilakukan swab diapatkan levofloksacin, moxifloxacin, obat
oropharyng dan nasopharyng PCR dua kali injeksi lini kedua sensitive. Hasil DST
dengan hasil negatif/positif. Pasien diberikan didapatkan resisten INH dan high dose INH,
antiviral dan OAT TB RO dilanjutkan. sensitive terhadap kanamisin, kapreomisin,
moxifloxacin, amikasin, dan pirazinamid. Pasien
Simpulan: Anamnesis, pemeriksaan fisisk dan
pennunjangdapat meniagnosis koinfeksi TB RO

dirawat di ruang isolasi dan dilakukan swab
oropharyng dan nasopharyng PCR dua kali
dengan hasil negatif/positif. Selama dirawat,
pasien diberikan antiviral (osletamifir 2xx75mg
selama 10 hari) dan Obat anti tuberkulosis TB
RO regimen individual. Pasien di rawat di RS
selama 10 hari, kemudian didapatkan perbaikan
klinis dan melanjutkan isolasi dirumah. Selama
isolasi dirumah pasien rutin mendapatkan obat
antituberkulosis dari puskesmas dan pasien
melakukan pemantauan mimum obat dengan
menggunakan video yang dikirimkan ke
puskesmas setiap hari. Seama penobatan Oat
pasien tiak didapatkan efek samping OAT.

DISKUSI
Pengalaman mengenai tatalaksana pasien
tuberkulosis dan covid-19 masih terbatas,
pemantauan akan outcome yang buruk harus
selalu diperhatikan. Tuberkulosis resisten obat
dan covid-19 memiliki gejala yang hampir sama
yaitu batuk, demam, dan sesak napas.
Pemberian dan pengawasan minum obat anti
tuberkulosis resisten obat merupakan tantangan
saat pasien melakukan isolasi mandiri.

SIMPULAN
Penatalaksanaan pada pasien ini diberikan
antiviral dan OAT regimen individual. Diagnosis
dan management yang tepat dapat mencegah
perburukan pada pasien TB RO dan covid-19.
Pemantauan pengobatan TB RO dilakukan
dengan pemanfaatan tehnologi digital yaitu
memakai video yang dikirimkan ke puskesmas
setiap hari.

DAFTAR PUSTAKA
1. Kumar R, Bhattacharya B, Meena V, Soneja

M, Wig N. Covid-19 and TB coinfection-
Finishing touch in perfect recipe to severity or
death. J infect 2020; 81(3):39-40.
2. Saunders MJ, Evans CA. Covi-19,
tuberculosis, and poverty: preventing a
perfect storm. Euripean Respiratory Journal
2020;Vol 57:3
3. Tadolini M, Garcia-garcia JM, Blanc FX,
Borisov S, Golleti D, Motta I, et al. On
tuberculosis and covid-19 coinfection.
Europian Respiratory Journal 2020;56:2.



PERBANDINGAN HASIL PEMERIKSAAN RADIOLOGIS (RONTGEN THORAX/ X RAY)
DAN HASIL LABORATORIUM (RAPID TEST ANTIBODI) PADA PASIEN

TERKONFIRMASI COVID 19 DI RS PARU DR. H.A ROTINSULU BANDUNG PERIODE
BULAN MARET S/D BULAN SEPTEMBER TAHUN 2020

Peneliti :
Dr. Edi Sampurno., SpP(K)., MM., FISR
dr. Sri Dyah Panji Kristiani Sukoco, Sp.Rad.

dr. Asima Manurung, M.K.K.K.
Rina., S.Si., Mkes

RS PARU Dr H A ROTINSULU BANDUNG
TAHUN 2020