Today marks the 1st anniversary of the very first COVID-19 patient being admitted to our hospital. It is hard to believe that it has been only a year since that date. So much has happened since then. But I certainly recall the deep foreboding that many of us in the medical field felt. It was a scary time since we knew absolutely nothing about this disease and yet we were seeing China, Italy, and Iran being overrun with sick patients. Seattle was also inundated with sick patients by then. We were also hearing about health care workers dying despite taking every precautions available. Naturally, we were yearning for information – anything at all – to give us a better sense of what we were going to confront: how sick were the patients?; how infectious was the virus?; how fast did patients clinically deteriorate?; what were some complications?; and what were the potential treatments?
One of the earliest sources of information was interestingly emails which were widely disseminated by front-line workers who had some working experiences with COVID-19. Through various connections (actually don’t recall who), I received several of these very informative emails and thought it would be useful at the time to summarize the findings and disseminate it the multiple hospital doctors across the Boston area. Here is the actual email. Surprisingly, in hindsight, the information contained here was actually pretty much on-the-mark, considering the dearth of data available on SARS-CoV-2 at the time.
______________________________________________________________________________________________________________________________
From: Ahn,Andrew
Sent: Thursday, March 12, 2020 11:54 PM
To:
Subject: Fw: covid experiences
I am forwarding 2 emails which contain information regarding the COVID frontline experience. One is from an intensivist in Seattle, while the other is from our own BIDMC Pulm division who had a Zoom conference earlier this afternoon with Wuhan intensivists. (Feel free to read emails below).
Based on all reports, we are experiencing similar trajectories as other regions in Europe (like Italy), and thus we should expect to have some pretty busy upcoming weeks-months..
From what I have been able to garner thus far, here is a quick summary of COVID facts/estimations:
- Estimated Incubation time (time of exposure to symptom development): median ~ 4-5 days (interquartile 2-7 days)
- Clinical Progression: Dyspnea (~6 days post-exposure), Admission (~8 days), ICU Admission/intubation (~10 days)
- Many start with “silent hypoxemia” (clinically benign but hypoxic) –> worsened dyspnea –> hypoxia and hypercarbic resp failure
- Physiologically, 2 stages
- Replicative Stage: viral replication with high amounts of shedding even before symptoms develop. Begin to decrease ~ 5 days post-exposure (but can last up to 10 days)
- Adaptive Immunity Stage: immune response which lead to decreasing viral titers, but when maladaptive responses can occur – “Cytokine storm” -> ARDS/DAD. Typically 7-10d post-exposure
- Hypotension uncommon
- Normal WBC, almost universally lymphopenic.
- Markers that may track with disease severity/prognosis: CRP, ferritin, LDH, D-dimer
- Port CXR are likely sufficient for assessing of the patient’s lung parenchyma
- Chest CT scans add little to management, overwhelms resources (need for transport, cleaning of CT scan suite, nursing effort, etc..), and increase risk of exposure
- Progression to resp failure is RAPID – reportedly in span of 12-24 hrs (usually begins 7-10 days post-exposure):
- Deterioration is quite fast and thus many are reporting that non-invasive ventilation (BiPAP/CPAP) are merely delaying the inevitable intubation and thus possibly worsening outcome
- Surprisingly high rates of cardiovascular deaths (VT/VF, asystole) attributed to a form of viral dilated cardiomyopathy
- Interestingly, seems to be occurring at or soon after recovery from respiratory failure.
- Reportedly EF can be normal on admission and at some point during admission, EF quickly drops to 10% in as quick as 12 hrs (anecdotally)
- May track with elevated troponin
- Steroids are likely more detrimental than helpful
- Consider d/c’ing prescribed inhaled steroids – as may diminish viral clearance. (unless there is clear clinical need [e.g., asthma exacerbation])
- Reflexive use of antibiotics not likely helpful. For those who utilize precalcitonin (like BWH/VA), it is usually low in COVID patients and can be monitored to assess presence of bacterial superinfection (a/w precalcitonin increases)
- No real antiviral treatments. Mixed reports on Remdesevir (which is not widely available on market anyhow)
- Clinical course is quite protracted. Not atypical for intubated patients to have overall LOS >= 14 days
Treatment considerations for Hospitalists:
- Evidence that greatest amount of viral shedding occurs in early stage of disease – many of these contagious patients will be our floor patients. Adequate resp protection is critical – AND because viral shedding can occur before sxs develop, it might be wise for us to don on new masks (not necessarily N95) after leaving the COVID room to prevent us from being the asymptomatic vectors of the virus to other patients/staff
- Do not rely much on BiPAP/NIV. If anything, it may worsen outcome. If respiratory detoriation occurs, recognize that decline is frequently rapid and would most benefit from early intubation (based on reports from both Seattle and Wuhan)
- Do NOT overtreat with fluids. By all reports, septic shock physiology is uncommon and delivery of too much fluid can worsen ARDS/DAD
- Do not be overly complacent with patients coming out of the ICU onto the floor. Reportedly, arrhythmia and severe dilated cardiomypathy can happen post resp-failure recovery.
- Be prepared for multiple EOL discussions. Because these admissions tend to have prolonged hospitalizations and long ICU stays, we must recognize that Mechanical ventilation, ICU beds, and other critical-level care may be quickly saturated. Teasing out those individuals with poor prognosis may be best for both his/her QOL and for the whole medical system
Potential HMED contingencies:
- By reports, BIDMC is capable of mechanically ventilating up to ~130 patients day. However, our anesthesia/pulm divisions are not that big – and with many of their staff having child-care needs and possibly contracting SARS-CoV2 themselves, HMED may need to serve as the critical care backup, being that our staff # is much bigger and some already have experiences in open ICUs at community sites. I know the idea of having us manage vent settings seems ridiculous, but in Milan, reportedly even pathologists have been recruited to take care of intubated COVID patients (not sure if actually true, but you get the point). It might be useful for us to select a few HMED staff to get re-educated on Vent-mgmt, paralytics, pressors, and vent weaning – and possibly have them educate others in HMED just in case this need were to ever to arise.
- With the risks for dilated CM, rapid cardiac detioriation, and logistical challenges of getting a TTE, it might be worth investing in a few portable ultrasounds and have us learn how to assess Cardiac EF, IVC dilation, and even pulmonary parenchyma at the bedside… This is a skill that HMED needs anyhow, and there will certainly be a need in these COVID patients.
- Maybe we should take Pulmonary’s example and contact Hospitalists in the Seattle area who may have important lessons to provide us on how they have managed the COVID cases thus far. May be really educational ….
If anyone has any additional thoughts, feel free to reply here.
Thanks,
Andrew
_________________________________________________________________________________________________________
From:
Sent: Thursday, March 12, 2020 3:13 PM
To: Ahn,Andrew
Subject: [External] FW: covid experience
The following description is being circulated among intensivists related to the Seattle experience as described by an intensivist. I cannot verify accuracy.
“We’ve been told not to share info, but we are all doing it anyway.
Since COVID is now deemed endemic in the XXXX area, and to quote a reliable source, the rest of the country is just “lagging behind,” thought I’d share some relevant details, including from CDC teleconference today for COVID providers.
– as we all assumed, it has been in community spread locally for weeks. We have seen idiopathic ARDS cases since early/mid-Feb. Retrospective testing is being done where possible. – the numbers presented in media do not reflect actual cases, obvs. Testing here only started 2/28. Our first CONFIRMED death was 2/23.
=XXX State Lab can only run 26kits/day, though they are ramping up quickly. Despite strict criteria for testing, there is a 3d backlog at this time.
– Negative Resp Path PCR is required before SARS2 test will be accepted. We have been running out of RP PCRs. This is unheard of, especially as most admitted resp pts get one during flu/cold season (mostly for approp iso, since RSV is contact). Goddess bless the local Children’s hospital for sending us 60 the other night. Your hospital should begin stocking up on RP PCRs now. Our Public Health dept does not expect SARS2 tests to be ample enough to d/c the neg RP PCR requirement.
– on a related note, county lab no longer runs tests from pts not sick enough to be admitted, since dz is now endemic. Expect this will be the case elsewhere soon.
– as of today, we have 21 pts and 11 deaths since 2/28. Not including the postmortem retrospective dx of pts who died with idiopathic ARDS the prior week. Of note, Harborview had an idiopath ARDS death 2/26. There will be more retrospective dx. – our mortality rate is skewed up (and in some cases, down) because many of our pts come from the LCCK SNF (Lifecare Care Center of Kirkland) & are elderly and severely chronically medically ill – the sort of pts who die of rhinovirus. Many of these patients’ families are opting for comfort care, as many are DNI. We have 3 such on the floor on comfort care now. Of note, those 3 pts have what would be considered mild infxn in a different cohort.
– we are seeing pts who are young (20s), fit, no comorbidities, critically ill. It does happen. – media (including NYT) are mentioning “efforts to contain the outbreak” at the SNF.
I’m sure you are all aware, but the US has been past containment since January, and the SNF cases aren’t an “outbreak” they’re a cluster. – thus far many pts have contacts there (esp visiting family members), but also at a local HD center and a car dealership. Others have zero identifiable contacts at all, tho I suspect many have Costco-horde connections, heh. – fortunately Evergreen has capability to turn all or half of any ward into a neg pressure zone
Currently, all of ICU is for critically ill COVIDs, all of XXX floor medsurg for stable COVIDs and EOL care, half of PCU, half of ER. New resp-sx pts Pulmonary Clinic offshoot is open.
– in XXXX, CDC is no longer imposing home quarantine on providers who were wearing only droplet iso PPE when intubating, suctioning, bronching, and in one case doing bloody neurosurgery on these pts in the week prior to testing starting. Because that resulted in our Stroke Center hospital no longer being able to admit LVOs or any kind of bleed. And decimated 10% of our Hospitalists, 3 of the 6 Night docs, and a PCCM. Plus it’s now endemic. Expect when it comes to your place you may initially have staff home-quarantined. Plan for this NOW. Consider wearing airborne iso PPE for aerosol-generating procedures in ANY pt in whom you suspect COVID, just to prevent the mass quarantines.
– we ran out of N95s (thanks, Costco hoarders) and are bleaching and re-using PAPRs, which is not the manufacturer’s recommendation. Not surprised on N95s as we use mostly CAPRs anyway, but still. Supplies are en route, but your facility may wish to stock up now, esp if you expect each staff member and room to have its own PAPR/CAPR.
– terminal cleans (inc UV light) for ER COVID rooms are taking forever, Enviro Services is overwhelmed. Bad as pts are stuck coughing in the waiting room. Rec planning now for Enviro upstaffing, or having a plan for sick pts to wait in their cars (that is not legal here, sadly).
– CLINICAL INFO based on our cases and info from CDC conf call today with other COVID providers in US:
– the Chinese data on 80% mildly ill, 14% hospital-ill, 6-8% critically ill are generally on the mark. Data very skewed by late and very limited testing, and the number of our elderly pts going to comfort care. – being young & healthy (zero medical problems) does not rule out becoming vented or dead – probably the time course to developing significant lower resp sx is about a week or longer (which also fits with timing of sick cases we started seeing here, after we all assumed it was endemic as of late Jan/early Feb). – based on our hospitalized cases (including the not formally diagnosed ones who are obviously COVID – it is quite clinically unique) about 1/3 have mild lower resp sx, need 1-5L NC. 1/3 are sicker, FM or NRB. 1/3 tubed with ARDS. Thus far, everyone is seeing: – nl WBC. Almost always lymphopenic, occasionally poly-predominant but with nl total WBC. Doesn’t change, even 10days in. – BAL lymphocytic despite blood lymphopenic (try not to bronch these pts; this data is from pre-testing time when we had several idiopathic ARDS cases) – fevers, often high, may be intermittent; persistently febrile, often for >10d. It isn’t the dexmed, it’s the SARS2. – low ProCalc; may be useful to check initially for later trending if later concern for VAP etc. – up AST/ALT, sometimes alk phos. Usually in 70-100 range. No fulminant hepatitis. Notably, in our small sample, higher transaminitis at admit (150-200) correlates with clinical deterioration and progression to ARDS. LFTs typically begin to bump in 2nd week of clinical course. – mild AKI (Cr <2). Uncertain if direct viral effect, but notably SARS2 RNA fragments have been identified in liver, kidneys, heart, and blood.
– characteristic CXR always bilateral patchy or reticular infiltrates, sometimes perihilar despite nl EF and volume down at presentation. At time of presentation may be subtle, but always present, even in our pts on chronic high dose steroids. NO effusions. CT is as expected, rarely mild mediastinal LAD, occ small effusions late in course which might be related to volume status/cap leak.
– Note – China is CT’ing everyone, even outpts, as a primarily diagnostic modality. However, in US/Europe, CT is rare, since findings are nonspecific, would not change management, and the ENTIRE scanner and room have to terminal-cleaned, which is just impossible in a busy hospital. Also, transport in PAPRs. Etc. 2 of our pts had CTs for idiopathic ARDS in the pre-test era; they looked like the CTs in the journal articles. Not more helpful than CXR. – when resp failure occurs, it is RAPID (likely 7-10d out from sx onset, but rapid progression from hospital admit). Common scenario for our pts is, admit 1L NC. Next 12hrs -> NPPV. Next 12-24hrs -> vent/proned/Flolan. – interestingly, despite some needing Flolan, the hypoxia is not as refractory as with H1N1. Quite different, and quite unique. Odd enough that you’d notice and say hmmm. – thus far many are dying of cardiac arrest rather than inability to ventilate/oxygenate. – given the inevitable rapid progression to ETT once resp decompensation begins, we and other hosps, including Wuhan, are doing early intubation. Facemask is fine, but if needing HFNC or NPPV just tube them. They definitely will need a tube anyway, & no point risking the aerosols. – no MOSF. There’s the mild AST/ALT elevation, maybe a small Cr bump, but no florid failure. except cardiomyopathy. – multiple pts here have had nl EF on formal Echo or POCUS at time of admit (or in a couple of cases EF 40ish, chronically). Also nl Tpn from ED. Then they get the horrible resp failure, sans sepsis or shock. Then they turn the corner, off Flolan, supined, vent weaning, looking good, never any pressor requirement. Then over 12hrs, newly cold, clamped, multiple-pressor shock that looks cardiogenic, EF 10% or less, then either VT->VF-> dead or PEA-> asystole in less than a day. Needless to say this is awful for families who had started to have hope. – We have actually had more asystole than VT, other facilities report more VT/VF, but same time course, a few days or a week after admit, around the time they’re turning the corner. This occurs on med-surg pts too; one today who is elderly and chronically ill but baseline EF preserved, newly hypoTN overnight, EF<10. Already no escalation, has since passed, So presumably there is a viral CM aspect, which presents later in the course of dz. – of note, no WMAs on Echo, RV preserved, Tpns don’t bump. Could be unrelated, but I’ve never seen anything like it before, esp in a pt who had been HD stable without sepsis.
Treatment –
– Remdesivir might work, some hosps have seen improvement with it quite rapidly, marked improvement in 1-3 days. ARDS trajectory is impressive with it, pts improve much more rapidly than expected in usual ARDS.
– Recommended course is 10d, but due to scarcity all hosps have stopped it when pt clinically out of the woods – none have continued >5d. It might cause LFT bump, but interestingly seem to bump (200s-ish) for a day or 2 after starting then rapidly back to normal – suggests not a primary toxic hepatitis.
– unfortunately, the Gilead compassionate use and trial programs require AST/ALT <5x normal, which is pretty much almost no actual COVID pts. Also CrCl>30, which is fine. CDC is working with Gilead to get LFT reqs changed now that we know this is a mild viral hepatitis.
-currently the Gilead trial is wrapping up, NIH trial still enrolling, some new trial soon to begin can’t remember where.
– steroids are up in the air. In China usual clinical practice for all ARDS is high dose methylpred. Thus, ALL of their pts have had high dose methylpred. Some question whether this practice increases mortality.
– it is likely that it increases seconday VAP/HAP. China has had a high rate of drug resistant GNR HAP/VAP and fungal pna in these pts, with resulting increases mortality. We have seen none, even in the earlier pts who were vented for >10d before being bronched (prior to test availability, again it is not a great idea to bronch these pts now).
– unclear whether VAP-prevention strategies are also different, but wouldn’t think so?
– Hong Kong is currently running an uncontrolled trial of HC 100IV Q8.
– general consensus here (in US among docs who have cared for COVID pts) is that steroids will do more harm than good, unless needed for other indications.
– many of our pts have COPD on ICS. Current consensus at Evergreen, after some observation & some clinical judgment, is to stop ICS if able, based on known data with other viral pneumonias and increased susceptibility to HAP. Thus far pts are tolerating that, no major issues with ventilating them that can’t be managed with vent changes. We also have quite a few on AE-COPD/asthma doses of methylpred, so will be interesting to see how they do.
That’s all I got for now. Will be skipping the next 2 CDC COVID calls as working Nights, but will call in again next week and keep you all posted.
Plz share info but preferably with no direct attribution as I need to remain employed”
Sent from my iPhone
______________________________________________________________________________________________________________________________
From:
Sent: Thursday, March 12, 2020 1:27 PM
To:
Subject: Phone Call with Wuhan Intensivists
Hello,
I wanted to share with you all the details of a Zoom meeting we had with some Wuhan intensive care physicians this morning. They seemed in better spirits than expected and were hopeful that they were starting to see things getting better. We really are part of a global community dealing with this together now, and it was nice to get their thoughts and suggestions. I have summarized the narrative below.
Thanks,
Overview of Discussion with Wuhan
1) Natural History: The course of coronavirus appears to be quite prolonged in the patients who end up critically ill. It can take 5-7 days from time of initial symptoms to develop respiratory distress and hypoxemia and about 10 days until patients require ICU admission. If patients are less severely ill and are ok on nasal cannula alone usually these patients get better over the next 10-14 days. If patients become more hypoxemic and require intubation, it can take 2-3 weeks for recovery. It appears the pathology while not fully fleshed out is consistent with DAD and typical ARDS in the patients who are mechanically ventilated. Patients initially can worsen in a “silent hypoxemia” phase where they do not look too bad clinically but have pretty severe V/Q mismatch and hypoxemia, and the early phase of the disease is more hypoxemia alone. As the disease progresses it becomes a mixed hypercapneic and hypoxemic failure as the dead space fraction worsens. While some patients can be quite stiff, this is not universally the case and some patients may have severe hypoxemia without significant decreases in compliance/increases in driving pressure. Development of AKI appeared common in the ICU population in 10-30% of patients some of whom require CVVH. The disease tends to spare hemodynamic effects without significant hypotension at least in the early phases (although this is somewhat in contrast to what we had heard from Italy). There is some concern for cardiac toxicity with roughly 30-50% of patients having elevations in troponins, and the presence of cardiac toxicity the intensivist thought correlated with mortality. Bacterial and fungal superinfection appears to be uncommon overall.
2) Treatment: In their experience early intubation seems to improve patient outcomes. Because of limited resources they were forced to use high flow in some patients for longer than they would have preferred and there was concern in these patients for increased transmission as well as suboptimal control and worsened outcomes. They found that proning appears to be very effective in the more critically ill patients and 70-80% of patients had a “good” response to proning in terms of oxygenation, and they utilized a similar approach with 16+ hours of proning per day. They also found they were able to avoid paralyzing patients more often if intubated earlier as the severity of their disease appeared to be less. In the “late” intubations, the use of paralytics was increased. They do not have inhaled NO in Wuhan, so were unable to comment on differences between veletri and iNO. There was differing opinions as to the utility of recruitment maneuvers and high PEEP levels between the Wuhan clinicians. The response to steroids is unclear. Although 60-70% of vented patients received steroids, there was unclear benefit from this, and the clinician we spoke to thought that longer term it may have resulted in more prolonged viral clearance and slower recovery even if patients appeared briefly to benefit initially. They have used several antiviral medications (including remdesevir), but did not feel like they could recommend their use and were unsure of their benefit. They suggested a conservative fluid management approach as most patients were relatively preserved from an HD perspective. Regarding initiation of ECMO, they found that patients did not seem to do better with early ECMO initiation, and they used it more in a salvage approach, but also used this approach because of limited resources. Most patients were treated up front empirically for CAP/HAP, but the benefit of this was unclear, and generally longer courses of antibiotics in most patients were not suggested.
3) Prevention: Overall health care worker protection has been important. Because of the numbers of patients they were unable to sequester patients to specific areas of the hospital as essentially the entire hospital became a COVID hospital. They routinely use N95 and face shields for all interactions with patients, but they mentioned that PPE use was variable across different hospitals. Additionally they barred all visitors from coming to prevent further spread. They also agreed with early intubation and limiting NIV and BiPAP as much as possible. They routinely use humidified circuits with a filter on the expiratory tube to limit spread in ventilated patients.