COVID-19: comorbidities and outcomes

blog from 4/4/20

2 articles came out of note on a more detailed analysis of the association between underlying health conditions and Covid-19, one from China and one from the US.

 

CHINA: see covid comorbidities and outcomes eurrespj2020 in dropbox, or doi: 10.1183/13993003.00547-2020

   

Details:

-- 1590 laboratory-confirmed hospitalized patients from 575 hospitals in 31 provinces across China

-- mean age 50, 43% female, severe cases in 16%

-- most common symptoms on or after hospitalization: fever 88%, dry cough 70%, fatigue 43%, productive cough 36%. Abnormal chest CTs in 70%

-- data analyzed between December 11, 2019 and January 31, 2020

-- composite endpoint: admission to ICU, invasive ventilation, or death

-- median follow-up: 10 days

 

Results:

-- composite endpoint: 131 patients (8%)

    -- death: 50 patients (6%)

    -- ICU: 99 patients (6%)

    -- invasive ventilation: 50 patients (3%)

-- at least one comorbidity: 399 patients (25%)

    -- nonsevere cases: 10% had at least one comorbidity

    -- severe cases: 33%

    -- age: mean of 61 (at least one comorbidity) vs 45 (no comorbidity)

    -- clinical presentation: those with at least one comorbidity had more shortness of breath (41% vs 18%), nausea or vomiting (10% vs 4%), and abnormal chest x-ray (29% vs 15%)

-- prevalent comorbidities, presence of severe vs non-severe cases:

    -- hypertension: 17%; 269 cases, 33% severe  vs 13% nonsevere

    -- diabetes: 8%; 130 cases, 35% severe vs 14% nonsevere

    -- other cardiovascular diseases: 4%. cerebrovascular disease, hepatitis B, COPD, C KD, malignancy, immunodeficiency: all less than 2%

    -- 2 or more comorbidities: 8%

-- adjusting for age and smoking status, presence of severe vs non-severe cases:

    -- COPD, HR 2.7 (1.4-5.0)

    -- diabetes, HR 1.6 (1.0-2.5)

    -- hypertension, HR 1.6 (1.1-2.3)

    -- malignancy, HR 3.5 (1.6-7.6)

-- number of comorbidities, presence of severe vs non-severe cases:

    -- at least one comorbidity, HR 1.8 (1.2-2.8)

    -- 2 or more comorbidities, HR 2.6 (1.6-4.2)

 

Commentary:

-- as a perspective, prior studies suggested that any comorbidity was associated with a 3.4-fold increased risk of developing ARDS in patients with influenza H7N9 infection; similar numbers with other influenzae, SARS, and MERS. These studies in general had relatively small sample sizes and single center observations

-- comorbidities were determined by patient self-report on admission; more than 1 in the same organ system (e.g. coronary heart disease and hypertension) would count as one comorbidity

-- for hypertension, the high prevalence and 60% increase in adjusted risk might be explained by several factors:

     --people might have had multiple comorbidities, but hypertension was the only one the patient knew about. For example, patients may not know that they have COPD (never had spirometry, but may have had significant smoking or occupational history and undocumented COPD; and my guess is it is unlikely to have such a low rate of COPD, 24 people/1.5% in the Chinese population living in an industrial region, as found in the study). Or, patients may have had more asymptomatic but severe coronary artery disease related to hypertension (and, perhaps smoking: no data provided; they did control for smoking, but not sure what that means: eg smoking 1 cigarette vs 3 packs per day poses a different risk)

    --the presence of comorbidities in general might be a surrogate for differences in frailty or ability to function more socially, and thereby change the dynamics of viral transmission from those without comrobidities; hypertension may just be a marker of underlying differences in social interactions (more dependent on others, more family contact…. And 90% of viral transmissions in Wuhan were related to family clusters)

    --and, this also brings up the rather controversial issue of ACE-2 levels (those are the receptors of SARS-CoV-2 in the lung, and some by not all studies suggest that the use of ACE-inhibitors, as often used in those with hypertension and diabetes, may increase ACE-2 levels (see http://gmodestmedblogs.blogspot.com/2020/03/covid-ace2-ibuprofen-and-grasping-for.html). But, as with much COVID-related, we have minimal granular information, including about meds being taken.

 

 

US: see https://www.cdc.gov/mmwr/volumes/69/wr/pdfs/mm6913e2-H.pdf

 

Details:

-- as of March 28, the US had 122,653 cases of Covid-19, and 2112 deaths

-- the data below is from laboratory-confirmed Covid-19 cases reported to CDC from 50 states, 4 US territories, District of Columbia, New York City (cases from people repatriated to the US from Wuhan and the Diamond Princess cruise ship were excluded)

-- in 7162 cases (5.8%) there is information on underlying health conditions or other known risk factors for severe outcomes from respiratory infections

    -- 2692 (38%) of these cases had one or more underlying health conditions or risk factors; 4470 (62%) had none of these conditions

-- for those missing illness onset dates, date of illness onset was estimated by subtracting 4 days from the earliest specimen collection

-- potential risk factors assessed: chronic lung disease, diabetes, cardiovascular disease, chronic renal disease, chronic liver disease, immunocompromised conditions, neurologic disorders, pregnancy, current smoking status, former smoking status, and "other chronic disease" (this included an array of problems, most quite uncommon, except for hypertension, with 113 cases)

-- only people greater than 19 years old were considered (sample size of younger people was inadequate)

-- to counter missing data, the following percentages were given a range

 

Results:

-- most commonly reported health conditions associated with increased risk:

    -- diabetes: 784 individuals, 11%

    -- chronic lung disease: 656 individuals, 9%

    -- cardiovascular disease 647 individuals, 9%

-- for those with at least one underlying health condition/risk factor, vs none:

    -- ICU admissions: overall percentage 13.3% - 14.5% vs 2.2% - 2.4%, 358 of 457 (78%) of the ICU admissions

    -- hospital admission without ICU: overall percentage 27.3%-29.8%, vs 7.2%-7.8%, 732 of 1037 (71%) of the non-ICU admissions

    -- those not hospitalized: 1388 of 5143 (27%)

-- deaths: 184 had complete information, 173 (94%) had at least one underlying condition

-- by age:

    --19-64 yo:

        -- hospitalized without ICU admission: 285 (18.1%-19.9%) had at least one underlying condition vs 197 (6.2%-6.7%) did not

        -- hospitalized with ICU admission: 134 (8.5%-9.4%) had at least one underlying condition vs 58 (1.8%-2.0%) did not

    -- >65yo:

        -- hospitalized without ICU admission: 425 (41.7%-44.5%) had an underlying condition vs 58 (16.8%-18.3%) did not [pretty low numbers of patients in those not with underlying condtions]

        -- hospitalized with ICU admission: 212 (20.8%-22.2%) at an underlying condition vs 20 (5.8%-6.3%) did not [pretty low numbers of patients in those not with underlying condtions]

 

Commentary:

-- unfortunately, in the midst of the pandemonium associated with the pandemic, we have relatively minimal data regarding Covid-19. It is quite notable that only 5.8% of Covid cases have any report on underlying health conditions. It is understandable, of course, that in the midst of the onslaught of severe illness, and the inadequacies of our health system in preparation for this onslaught (PPEs, respirators, hospital beds), that few clinicians are spending time documenting epidemiology of the virus

-- one issue with this data is that a very large percentage of the United States population has at least one of these conditions, especially diabetes in 10% of the population) and hypertension (about 30% of the population). And these numbers increase dramatically with age: for example in the Framingham study, >80% of those >60 yo had hypertension

-- though there were very small numbers and missing data, this is the 1^st study I’ve seen that disaggregated age from comorbidities. As we all know, aging is associated with increasing comorbidities, so it is useful to know whether a very healthy 75-year-old is in fact at increased risk (e.g. as compared to a 40-year-old with comorbidities). this study did suggest that both ICU admissions and hospitalizations without ICU admissions were about 3-fold lower in healthy people >65yo, and were roughly equivalent to those aged 19-64 with an underlying condition

-- this data is also biased by the fact that in the US we did not have nearly sufficient SARS-CoV-2 testing equipment, so the cases in the US were all much sicker people who met the strict testing criteria, and missed the likely vast majority who had mild symptoms that did not reach that bar (or the asymptomatics...)

-- also, it is not clear that the data are entirely consistent: the criteria for reporting underlying conditions to the CDC were not explicit and likely varied from place to place in the country, institution to institution, individual clinician from one to the other.

-- And the amalgamation of conditions put into broad categories is problematic:  the category"chronic lung diseases" likely has diseases with different risk (e.g. COPDmay be worse than some other pulmonary conditions??)

 

So, it is not surprising that there is global consistency of the results suggesting that comorbid conditions increase likelihood of more severe Covid infections (also true in Italy and other areas). The US study, though severely limited in its data collection and reporting, does suggest that medical comorbidities are likely more important than age itself as a prognosticator (not that we should not be protective of our dear elderly based on this limited data). It would be very useful to have a much larger and more consistent database, and include a large sample of people who have mild-to-moderate as well as asymptomatic infections.

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