COVID: Italy's ICU outcomes; end-of-life care

a recent article evaluated COVID-positive patients admitted to the ICU in northern Italy (see covid italy icu admits jama2020 in dropbox, or doi:10.1001/jama.2020.5394). thanks to my brother Andrew for bringing this to my attention

Details:
--1593 (9%) of the 17,713 Covid-positive people in Lombardy, Italy were admitted to the ICU

--82% males (similar sex distribution in different age groups)

--age distribution: median age 63, with about equal percentages of those 54-58yo as 70-74yo, and 46-50yo as 74-78yo (ie, only a mildly skewed bell-shaped curve centering on age 62-66)

--comorbidities (present in 68%): hypertension 49%, cardiovascular disease (including cardiomyopathy and heart failure) 21%, hypercholesterolemia 18%, diabetes 17%, malignancy 8%, COPD/CKD/chr liver dz all <5%. not surprisingly, comorbidities increased with age

--respiratory support in 1300, 88% intubated and receiving invasive mechanical ventilation, 11% noninvasive ventilation

--PaO₂/FiO₂ median 160, with the single highest group (33%) having values of 75-125  (ie: really bad)

Results:

--see the article for many of the ICU details, such as PEEP levels (which got really high for some)

--most of the 1581 patients (920 of them) were still in the ICU at the time the data were collected. 256 had been discharged and 405 had died

--ICU mortality increased with age: 15% of those <63yo, vs 36% of those >63yo.  those 71-80yo had mortality rate of 40%. too few admitted to ICU >80yo (22 people) to have meaningful data (though 11 died)

    --stratifying by hypertension: overall 38% died vs 22% without hypertension (though those who were hypertensive were also older: 66 vs 62 yo)

--ICU length of stay: median 9 days (6-13). those still in the ICU had been there a median of 10 days. those discharged 8 days, and those who died 7 days

Commentary:

--the Italian experience is undoubtedly different from elsewhere, limiting its generalizability: different thresholds to admit to the ICU, different criteria for admission (eg: can people elsewhere get noninvasive ventilation on the wards, as they can in Italy?), different resources available (including ICU beds, ventilators), the volume of sick patients at a time who would qualify for admission (leading to prioritizing admissions), etc. However, some notable differences were found in Italy vs other areas:

    --invasive mechanical ventilation in Italy was in 88% (they did have huge numbers of sick people at once, coupled with inadequate resources, and were doing non-ICU noninvasive ventilation)

        --other areas: Seattle 71% (19% got noninvasive ventilation); China: 30-47% (42-62% got nonivasive ventilation) [of note, in China 1/2 were on ECMO, only 5 patients got that in Italy]

    --mortality rates in those requiring ICU admission varied from area to area: i could not find good studies on this, though early data had very high rates in small numbers of patients (eg, in China it was reported in the 80%+ range; early data from Seattle on small numbers of patients was on the order of 50% in https://newsroom.uw.edu/news/early-study-covid-19-patients-shows-high-mortality-rate ,or covid seattle ICU pts outcome NEJM2020 in dropbox or DOI: 10.1056/NEJMoa2004500). New York, per Gov Cuomo: "only 20% of coronavirus patients placed on ventilators will ever come off": se https://www.usatoday.com/story/news/health/2020/04/08/coronavirus-cases-ventilators-covid-19/2950167001/ .  a real dearth of robust data....  though, this Italian study did find that the mortality rate for older people was on the order of 40-50%

-- it was surprising that 82% of ICU admissions in Italy were males. other data do suggest that males do less well, but 82% is higher than in other places. ?if any selection bias in testing/case ascertainment/reporting??

--the real reason i am bringing up this article is that i think it really raises questions about how we should discuss COVID with patients/families, especially since there is no specific therapy for Covid-19. realistically, severe COVID leads to several potential problems, and many patients/families may not understand this:

    --the overall length of hospital stay varies in ICU-bound patients, but is often in the 2-3+ week range

    --admission criteria to the ICU varies a lot from area to area, but at this point it still seems that there is a pretty high mortality rate in those on mechanical ventilation, which is done in a majority of ICU admissions

    --the mechanical ventilation itself is likely to last a pretty long time (average time in Seattle on mechanical ventilation for COVID was 10 days, vs usual of 3-7 days)

    --mortality is particularly high in older people and those with comorbidities

    --family members are likely not to be able to be there with their sick relatives: ie, through this often prolonged and pretty highly fatal ICU stay, patients often cannot be seen by family members (which is often very hard on both the patient and family)

--and, there are real concerns about the ability of many families to take care of Covid patients at home:

    --small living quarters shared by many people and the inability to effectively quarantine/self-distance (in China, 90% of transmission was from "family clusters")

    --inadequate knowledge/understanding of personal protection

    --inadequate PPE availability to help prevent viral spread

    --inadequate home-based support by health care workers

    --inadequate understanding of the trajectory of infection (eg that those more severely infected can see improvement followed by dramatic and rapid deterioration)

    --inadequate medications (eg, the run on albuterol inhalers may limit availability for those needing them, eg those with COPD/asthma)

    --and some with existing VNA support may not be able to continue or get the level of increased support they may need by nursing and home health aides.

--and, as per all issues Covid: we continue to have limited clinical information, in part because the health care workers are often way too busy/immersed in direct patient care that data are not collected consistently; and in part because of the woeful lack of community data on Covid infections:

    --it would be nice to be able to do rapid SARS-CoV-2 testing on all contacts (and those who are positive might be good candidates to attend to home-based care of a sick relative); or, even better, accurate rapid antibody testing to have those with apparently completed infections be able to attend to the sick

    --and, unfortunately with this virus, it seems to be the most contageous early, in those with minimal prodromal symptoms (or even asymptomatic). and we have no ability to test these people given our dismal lack of testing supplies (and, despite assurances for many weeks about the soon-to-be dramatic increases in availability, this has yet to materialize). a huge issue for the US is the dramatically increasing number of known cases, and evidence that, perhaps related to inadequacy of supplies (and perhaps from partisan politics), many likely cases of Covid are not ascertained or reported: see https://www.nytimes.com/2020/04/05/us/coronavirus-deaths-undercount.html  

so, an important study overall, especially in reinforcing our need as clinicians to be very upfront with patients and families about the current situation regarding hospitalization and ICU care. Of course, we need to assess the values and desires of the patients themselves after understanding the current realities (limited as they are by are meager data) and respect their desires. But it is important that they understand some of the complexities, including the likelihood of a prolonged ICU stay, likely without the benefit of the company of friends and family. and that it seems the likely benefit of ICU stay/mechanical ventilation diminishes significantly by age and existing comorbidities

geoff​

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

to get access to all of the blogs:

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list