key: cord-0999442-zfa0rv8o
authors: Fattouch, Khalil; Corrao, Salvatore; Augugliaro, Ettore; Minacapelli, Alberto; Nogara, Angela; Zambelli, Giulia; Argano, Christiano; Moscarelli, Marco
title: Cardiac surgery outcomes in patients with COVID-19: a case-series report.
date: 2020-10-22
journal: J Thorac Cardiovasc Surg
DOI: 10.1016/j.jtcvs.2020.09.138
sha: 37180bb1d9c849a73023fdeedeebf3ff7f24e918
doc_id: 999442
cord_uid: zfa0rv8o

Objective The impact of COVID-19 on the post-operative course of patients after cardiac surgery is unknown. We experienced a major Sars-CoV-2 outbreak in our cardiac surgery unit, with several patients that tested positive early after surgery. Here we describe the characteristics, the post-operative course and laboratory findings of these patients, along with the fate of the health-care workers. We also discuss how we reorganize and reallocate hospital resources in order to resume the surgical activity without further positive patients. Methods After diagnosis of the first symptomatic patient, surgery was suspended. Nasopharyngeal swabs were carried out in all patients and health care workers. Sars-CoV-2 positive patients were isolated and monitored throughout the in-hospital stay and followed-up after discharged until death or clinical recovery. Results Twenty patients were found positive to Sars-CoV-2 sometimes after cardiac surgery (mean age 69±10.4 y/o; median EuroSCORE-II 3(IQR 5.1)); median time from surgery to diagnosis was 15 days (IQR 11). Among them eighteen had received cardiac surgery and two of them transcatheter aortic valve replacement. Overall mortality was 15%. Specific COVID-19 related symptoms were identified in 7 patients (35%). Among the twelve health-care workers infected, one developed a bilateral mild-grade interstitial pneumonia. Conclusion COVID-19 infection after cardiac surgery, regardless the time of the onset, is a serious condition. The systemic inflammatory state that follows the extracorporeal circulation may mask the typical COVID-19 laboratory findings, making the diagnosis more problematic. A capillary reorganization of the hospital resources is necessary in order to safely resume the cardiac surgical activity.

The pandemic infection caused by the novel coronavirus 2 (Sars-CoV-2 / 103 COVID-19) has dramatically affected cardiac surgery worldwide 1 . Many cardiac 104 surgery centers were reconverted to COVID-19 units 2 . Consequentially, a significant 105 reduction of surgical cases has been reported, with a potential negative impact to the 106 community 3 . COVID-19 may be a lethal condition due to respiratory tract 107 involvement, and might represent a major contra-indication to surgery unless 108 emergent serious conditions (i.e. acute aortic dissection) arise 4 . 109

Little data is available on the effect of COVID-19 in patients undergoing 110 cardiac surgery procedures 5 . There are also no firm recommendations available to 111 guide decision making for patients requiring cardiac surgery during pandemic. 112

While assuring the treatment of urgent cases, we experienced a Sars-CoV-2 113 outbreak in our cardiovascular division (GVM Care&Research, Maria Eleonora 114 Hospital, Palermo, Italy) during the apex of the pandemic in Italy, with twenty 115 patients found positive with Sars-CoV-2 sometimes after surgery. Twelve health-care 116 workers were also affected. 117

The aim of this paper was to describe the baseline characteristics, the post-118 operative course and laboratory findings of these patients. In addition, measures 119

Italy, it was designated by local authority as one of the main "COVID-19 free" hub 128 for urgent and emergent cardiovascular procedures in the district of Palermo. At the 129 climax of the Italian outbreak in early April we were performing two to three major 130 cardiac operations each day. Patients were referred to us and accepted for surgery if 131 affected by non deferrable cardiac conditions such as acute coronary syndromes and 132 symptomatic coronary artery disease with unfavorable anatomy and/or not amenable 133 to percutaneous coronary intervention (i.e. left main stem, proximal left anterior 134 descending artery); symptomatic aortic valve stenosis or regurgitation or symptomatic 135 mitral valve regurgitation despite optimized medical therapy; disease of the ascending 136 aorta (symptomatic aortic aneurysm or dissection) and other short term life-137 threatening cardiac disorders (i.e. cardiac valves endocarditis). 138

Before hospital admission patients were screened for COVID-19 according to 139 the regional health care system recommendations, which included core temperature > 140 37°C, cough and/or dyspnea without a clear alternative cause, recent 141 domestic/international travel, contact with a known or suspected COVID-19 patient. 142

Nasopharyngeal swabs were not carried out, nor did we use fast kit to detect IgM and 143

IgG in the blood of both patients and health-care workers during the triage phase. Follow-up was obtained by directly contacting the patients themselves or the 157 COVID-19 units where patients were transferred or the patients themselves. All 158 patients were contacted by phone and they verbally expressed the will to participate to 159 the study. The study protocol was reviewed and approved by our internal Institutional 160

Review Board and the requirement for patient consent was waived. 161

There were fifty-five in-patients at the time of the COVID-19 outbreak in our 164 hospital. Among them, twenty patients were found positive to the Sarv-CoV-2 165 sometimes after surgery (mean age 71±11.5 y/o; of theme11 were female); eighteen 166 received cardiac surgery and two of them transcatheter aortic valve replacement 167 (TAVR); all patients were found positive, except one, while in-hospital, in the cardiac 168 or rehab ward. Table 1 reports the surgical patients' baseline characteristic and Table  169 2 the operative details. All the individuals who tested positive for the new coronavirus 170

were operated between March 9 th 2020 and April 06 th 2020, but one had the operation 171 on January 23 th 2020. Immediately after detection of the first positive patient, cardiac surgery 191 activity was suspended. All patients and health-workers were tested for Sars- Pre-operative negative patients were diverted to other hospitals. All the positive post-193 operative patients were transferred over to dedicated hemodynamically and respiratory stable. 195 dysfunction followed by multi-organ failure. He was diagnosed with COVID-19 the 203 day after surgery. One patient died two days after having been transferred over to 204 dedicated COVID-19 unit for severe respiratory and renal failure. He was a severely 205 critically ill patient with previous kidney transplantation and end stage renal disease, 206 who underwent emergent aortic valve replacement after having been diagnosed with 207 bacterial aortic valve endocarditis. 208

Rest of post-operative complication is described in Table 2 . Median time from 209 surgery to COVID-19 diagnosis was 15 days (inter quartile range (IQR) 11), 210 minimum 2 and maximum 76 days after surgery (Table 3) . Eighteen patients were 211

found positive before 30 th days from surgery; one 33 days and another one 76 days 212 after surgery. Seven patients (38.8%) showed typical symptoms that could be 213 associated with COVID-19 (Table 3 ). There were no complications that could be 214 Table 4 reports relevant available laboratory findings from baseline to the day 220 of COVID-19 diagnosis. There was a trend toward a reduction of lymphocytes count, 221 while neutrophils were over expressed from early after surgery (figure 2). The 222 antithrombin (AT) III value was toward the lower limit (mean 80.8±22.2%), while D-223

Dimer and fibrinogen were significantly elevated at the time of the COVID-19 224

The seventeen survivors (fifteen surgical patients and the two patients who 228 received TAVR) were followed-up on a regular basis by contacting the COVID-19 229 centers where they were transferred, or speaking to them by phone if they had been 230 discharged at home. Mean follow-up was 62.8±17.8 days, 100% completed. Nine 231 patients out of the seventeen survivors were transferred to quarantine hotels since they 232 did not need active medical treatment. later diagnosed with COVID-19 that died after surgery. In these two cases RT-PCR 275 tests was not performed pre-operatively.

The need for prolonged mechanical ventilation support may cause sternal 277 wound dehiscence after cardiac surgery. In our series one patient died as a result of a 278 serious bleeding from the dehisced sternal wound, while in isolation. 279

The dynamic changes of lymphocyte subsets of patients with novel 280 coronavirus disease and their correlation with the disease severity is still under 281 investigation 6, 7 . It has been hypothesized that the degree of lymphopenia (and pro-282 inflammatory cytokine storm) is generally higher in severe COVID-19 patients than 283 in mild cases 7, 8 . Some other evidences suggest that the neutrophil count is also higher 284 in patients with severe COVID-19 disease 6 . 285

In our series, among the three patients who died, a certain degree of pre-286 operative lymphopenia was observed in two patients (1.5 and 1.3 10 3 /mm 3 287 respectively) and severe in one (0.3 10 3 /mm 3 ). The following full blood count 288 analysis, demonstrated a persistent lymphopenia and significant higher count of 289 neutrophils. However, higher count of neutrophils could be observed in all positive 290 patients. Following extracorporeal circulation and ECMO, substantial decreases in the 291 number and function of some populations of lymphocytes is commonplace 9-11 . Others 292 have reported increases in the number of neutrophils 11, 12 and D-dimer 13 and reduced 293 AT-III level 13 . It is also known that the extracorporeal circulation may up-regulate the 294 IL-6 14-16 . In summary the extracorporeal circulation elicits an important systemic 295 inflammatory response syndrome (SIRS) 17, 18 that may last up to several days after 296 the index surgery 16 . Thereby, we believe that the extracorporeal circulation may act as 297 'confounder' and in some cases may delay the COVID-19 diagnosis and can 298 exacerbate the clinical scenario ( figure 3) . The SIRS state after extracorporeal 299 circulation may in some cases lead to acute respiratory acute respiratory distress 300 syndrome 19 . 301 J o u r n a l P r e -p r o o f Nonetheless, CT chest differential diagnosis may be an actual issue in patients 302 affected by COVID-19 after cardiac surgery. 303

At a consistent but slow pace, the cardiac surgery community is trying 304 worldwide to resume the surgical activity 20 . It is highly debated whether a second 305 wave of infection with a new pandemic will occur at certain point in the future. Many 306 urgent patients were denied cardiac surgery and cardiac interventions during the first 307 peak of the pandemic; also because at that time personal protection equipment and 308 nasopharyngeal swabs facilities were not freely available and accessible. Italy has 309 been among the first country to be severely hit by the COVID-19; however effort was 310 taken in order to provide, at least in each province, cardiac surgery coverage for 311 urgent and non-deferrable cases. While we were designed as dedicated 'COVID-free' 312 cardiac unit and we strictly followed the recommendations issued by the local 313 government (epidemiological screening, temperature screening, no visitors allowed), 314 and worn personal protection equipment, we could not avoid a Sars-Cov-2 outbreak, 315 among patients and health-workers. Intuitively the recommendations issued by the 316 local Italian authorities were insufficient to protect both patients and the health-317

workers community, yet they should be set in the context of an unprecedented global 318 shortage of nasopharyngeal swab and also personal protection equipment. 319

At the moment of the writing of this manuscript, the hospital policy has 320 radically changed. All elective patients are screened with nasopharyngeal swab before 321 admissions in to the clinical area with the possibility to have results ready in less than 322 an hour for the semi-elective or urgent cases. As others, we have created a dedicated 323 area in the hospital (named 'grey area' or 'bubble') (supplementary figure 1) , where 324 patients may stay until the nasopharyngeal swabs are processed. In such area face 325

shield and personal protection equipment as such as FFP2/3 (Filtering Face has dedicated resources such as echo-and ECG-machines and portable chest-X-ray 328 station with no cross-over with other clinical area. The 'bubble' has also one 'red-329 room' for urgent cases that may require invasive mechanical ventilation. 330

A no-visitors policy has been instituted since the beginning of the Sars-Cov-2 331 peak; 'essential visitors' are allowed in to the clinical area if they can provide a 332 negative recent (< 2 days) nasopharyngeal swab. All health-care workers undergo 333 nasopharyngeal swab every two weeks. 334

In supplementary figure 1 is presented the flow-chart on how we deal with 335 patients (urgent and elective or semi-elective) with COVID-19 status unknown along 336 with the 'bubble principles'. 337

What we have also learned from our experience is that after cardiac surgery, 338 COVID-19 diagnosis is not always intuitive. Suspicious respiratory disease after 339 cardiac surgery should be immediately investigated with nasopharyngeal swab and, in 340 some circumstances, specimen from the lower respiratory tract might be also tested 341

for Sars-Cov-2 (Supplementary table 2) . 342

In the light of a possible new pandemic, we believe that such hospital assets, 343 may guarantee a reasonable margin of safety. Our surgical activity (elective / semi-344 elective and urgent) resumed on July 2020; since then, following the described 345 screening measures, no further positive cases for Sars-Cov-2 were detected. 346

There are no firm recommendations available to guide decision making for 347 patients requiring cardiac surgery during pandemic 21 ; at this time of crisis, taking in 348 consideration also the possibility of a Sars-Cov-2 surge, it is pivotal to share 349 information and promote cooperation between associations of cardiac surgeons 21 .

This study has limitations. Perhaps the biggest weakness is that we could not 351 determine the date of the exact onset of the infection. Some patients may have been 352 exposed to the virus before surgery, yet it would be more rational to suppose that the 353 majority of them became infected at some stage after the operation. Finally we did not 354 explore correlation between severity of COVID-19 symptoms and risk factors for 355 cardiovascular diseases. 356

Here we presented a series of patients who underwent complex cardiac 359 surgery and were found positive for the new coronavirus after surgery. We could not 360 determine precisely when they were exposed to the Sars-CoV-2, however the Table 4 ) 377 378 Figure 3 . The impact of Sars-Cov-2 infection in patients referred for urgent cardiac 379 surgery is largely unknown. We experienced a major Sars-CoV-2 outbreak in our 380 cardiac surgery unit, with 18 patients that tested positive early after surgery and 2 381 after TAVI. 

Adult 391 cardiac surgery during the COVID-19 Pandemic: A Tiered Patient Triage Guidance 392 Statement

Society for Cardiac Surgery Task Force on C-P. The COVID-19 outbreak and its 13

Activation 426 of coagulation and fibrinolysis during cardiothoracic operations

IL-6 and endothelin-1 levels in patients undergoing coronary artery bypass 430 grafting. Do preoperative inflammatory parameters predict early graft occlusion and 431 late cardiovascular events?

Systemic inflammation after on-pump and off-pump coronary bypass surgery: a 436 one-month follow-up

The 438 inflammatory response to cardiopulmonary bypass: part 1--mechanisms of 439 pathogenesis

COVID-19 and ECMO: the interplay between coagulation and inflammation-a 442 narrative review

A nationwide 450 survey of UK cardiac surgeons' view on clinical decision making during the COVID-451 19 pandemic

Mitral valve regurgitation, n (%)

EuroSCORE 2, median (IQR) 3 (5.1)

Data are presented as mean ± standard deviation (SD) or median and interquartile range (IQR) or number and frequency (%)

NYHA: New York Heart Association; NSTEMI: Non ST elevation MI

Data are presented as mean ± standard deviation (SD) or median and interquartile range (IQR) or number and frequency (%). Acute kidney injury accordingly to RIFLE criteria

Baseline' vs 'Swab+', p=0.21; Lymphocytes: 'Baseline' vs 'Swab+', p<0.01; Neutrophils: 'Baseline' vs 'Swab+

Antithrombin III; CRP: C-reactive protein; NP: Nasopharyngeal; WBC: White cell count