Abstract
Objectives
Describe and evaluate the outcome of a coronavirus disease-2019 (COVID-19) patient without shortness of breath.
Design and methods
We retrospectively collected data from COVID-19 patients diagnosed and cared for in Marseille, France. We selected data from patients who at admission, had a low dose CT scanner, dyspnea status, and oxygen saturation available. Blood gas was analyzed in a sample subset of patients.
Results
Among 1712 patients with COVID-19, we report that 1107 (64.7%) do not complain of shortness of breath at admission. The low-dose computed tomography (LDCT) scan showed signs compatible with pneumonia in 757/1,107 (68.4%) of patients without dyspnea. In a subset of patients who had underwent at least one blood gas analysis (n = 161) and presented without dyspnea at admission, 28.1% (27/96) presented with a hypoxemia/hypocapnia syndrome. Asymptomatic hypoxia was associated with a very poor outcome (33.3% were transferred to the ICU and 25.9% died).
Conclusion
The absence of shortness of breath in an old patient with comorbidity merit medical attention and should not be considered as a good sign of well-being. The poor prognosis of asymptomatic hypoxia, highlight the severity of this mild clinical presentation. In these patients, pulse oximetry is an important mean to predict the outcome along with news score and LDCT scanner.
Keywords: COVID-19, Hypoxemia, Hypoxia, Hypocapnia, Happy, Low dose CT-scan, D-dimers, 02 sat
Introduction
Recently, a new coronavirus named severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) emerged in China at the end of December 2019 and rapidly spread throughout the world, producing millions of victims and several hundred thousand deaths (Huang et al., 2020). The early descriptive reports of the clinical presentation of the disease resulting from SARS-CoV-2 infection, named coronavirus disease-2019 (COVID-19), revealed that one third of patients did not have dyspnea (Zhou et al., 2020). Shortness of breath has been reported in 18.7% of 1,099 patients hospitalized with COVID-19, many of whom showed an abnormal CT scan (86%) and received supplemental oxygen (41%) (Guan et al., 2020). In patients with objective radiographic findings consistent with COVID-19 pneumonia, only 50% report shortness of breath (Huang et al., 2020). Despite this accumulating knowledge, the criteria for clinical screening and care management in most countries, including France, are still based on 3 symptoms: fever, cough, and dyspnea. Asymptomatic patients are requested to self-isolate at home, while those that become symptomatic are invited to contact their health care provider (Anon, 2020). However, some patients with COVID-19 deteriorate rapidly and seemingly without warning. For example, in Marseille, France, more than two-thirds of patients hospitalized in intensive care units (ICUs) come directly from home or are admitted to the ICU after less than three days of standard ward hospitalization. Additionally, many patients who later develop respiratory failure experienced hypoxemia and hypocapnia without signs of respiratory distress, particularly elderly patients. This is called "happy hypoxemia" or “silent hypoxemia” and was previously described in patients during the initial Wuhan outbreak (Xie et al., 2020). The cause of this asymptomatic hypoxia is not yet clear. Anosmia-hyposmia has been reported as a frequent clinical sign in COVID (Boudjema et al., 2020), but whether SARS-CoV-2 has access to the brain and contributes to the association between dyspnea and anosmia-hyposmia remain to be determined (Tobin et al., 2020). Some have links of happy hypoxemia with the development of thrombi within the pulmonary vasculature (Couzin-Frankel, 2020).
The discrepancy between respiratory clinical signs and the pulmonary lesions observed with a low-dose computed tomography (LDCT) scanner were reported in the early epidemic by Chinese authors (Huang et al., 2020, Zhou et al., 2020, Guan and Xian, 2020) but was not taken into account, and decisions based on home interviews of patients only considered dyspnea to justify hospital care. Massive test implementation and careful observation of positive patients, regardless of whether they were apparently symptomatic, allowed us to show that some patients who were classified as asymptomatic actually had hypoxia and others had lung damage visible on CT scan (Lagier et al., 2020). In this work, which is derived from a report based on a cohort of 3,737 patients, we wanted to focus on the relationship between dyspnea, hypoxemia, and lung lesions identified in LDCT scans.
Patients and methods
Patients
We retrospectively collected data from the information system of our hospital related to COVID-19 patients diagnosed and cared for in the infectious disease institute (IHU Mediterranée Infection) in Marseille, France, from March 3rd to April 27th. All the patients were diagnosed by PCR of a nasopharyngeal sample and were over 18 years of age (Lagier et al., 2020, La et al., 2020). We selected data from patients who had in their record at admission, all of the following items: a low dose CT Scanner, dyspnea status, and oxygen saturation available (1,712 patients). Blood gases were analyzed in a sample subset (161 patients). Dyspnea is a subjective sign, which was recorded at interview and was defined as shortness of breath. O2 desaturation was defined as SaO2< = 95%. A low-dose CT scanner was proposed for all patients as soon as possible after admission when possible, and radiological lesions were classified into four categories: normal, minimal grade 1, intermediate grade 2, or severe grade 3 (Liao et al., 2020). Hypoxemia was defined as pO2_mmHg< = 80, and hypocapnia as pCO2_mmHg< = 35. Demographics, chronic conditions (cancer, diabetes mellitus, chronic heart disease, hypertension, chronic respiratory disease, and obesity), concomitant medication, and signs and symptoms, including fever, cough, anosmia, ageusia, dyspnea, oxygen saturation, and blood gas analysis were extracted from the hospital information system (Lagier et al., 2020). Severity was assessed using the national early warning score adapted to COVID-19 patients (NEWS-2) (Liao et al., 2020).
Statistics
Categorical variables are presented as n (%). We used chi-squared test, Fisher's exact test, and univariate logistic regressions and estimated odds ratios (ORs) with 95% confidence intervals (CIs) to compare differences between patients with dyspnea and patients without dyspnea. To compare the prognosis performance of the LDCT scanner score and O2 saturation on clinical outcomes (transfer to ICU/death) among patients with no dyspnea, we performed multivariate logistic regressions. Models were adjusted on the following covariates: age, sex, time from symptom onset to admission, comorbidities (cancer, diabetes mellitus, cardiac diseases, hypertension, chronic respiratory diseases, and obesity), and clinical classification at inclusion (NEWS score). Four separate models were carried out. In the reference model (model A), all covariates previously listed were included. In Model B, we added to this list the LDCT scanner severity score. In model C, we added O2 saturation. Model D included both O2 saturation and the LDCT scanner severity. We calculated the Akaike information criterion (AIC) and computed the C-statistic for each model. The C-statistic is equal to the area under the receiver-operating characteristic (ROC) curve and ranges from 0.5 to 1. A two-sided α value of less than 0·05 was considered statistically significant. To investigate associations between clinical data, biological data, radiological data, and clinical outcomes (death and/or ICU stay), we performed univariate logistic regressions and principal component analysis (PCA) in a subset of 161 patients who had undergone at least one blood gas analysis. Hierarchical clustering on principal components (HCPC) was performed to identify clusters of patients. Analyses were carried out using the SAS 9.4 statistical software (SAS Institute, Cary, NC). PCA and HCPC were performed by using the R Statistical Software and the FactoMineR package (Lê et al., 2008).
Results
Among 3,737 patients diagnosed with SARS-CoV-2 infection in our institute, we selected 1,712 with available dyspnea, O2 saturation, and LDCT scan information obtained within the first 48 or 72 h. Among them, 1,107/1,712 (64.7%) presented with no dyspnea, and 605/1,712 (35.3%) presented with dyspnea. Underlying conditions and clinical symptoms are comprehensively described in Table 1 . The prevalence of poor clinical outcome (transfer to intensive care unit and/or death) significantly increased among patients with dyspnea.
Table 1.
Clinical characteristics of patients according to dyspnea status (n = 1,712).
| No dyspneaa (n = 1107, 65%) |
Dyspneaa (n = 605, 35%) |
p-valueb | All (n = 1712) |
||||
|---|---|---|---|---|---|---|---|
| n | % | n | % | n | % | ||
| Sex | |||||||
| Men | 529 | 47.8 | 257 | 42.5 | 0.035 | 786 | 45.9 |
| Age at inclusion | |||||||
| <45 y.o | 361 | 32.6 | 203 | 33.6 | 0.314 | 564 | 32.9 |
| 45−54 y.o | 273 | 24.7 | 157 | 26.0 | 430 | 25.1 | |
| 55−64 y.o | 239 | 21.6 | 143 | 23.6 | 382 | 22.3 | |
| 65−74 y.o | 122 | 11.0 | 53 | 8.8 | 175 | 10.2 | |
| ≥75 y.o | 112 | 10.1 | 49 | 8.1 | 161 | 9.4 | |
| Time from symptom onset to admission | |||||||
| <3 days (or no symptom) | 234 | 21.1 | 59 | 9.8 | <0.001 | 293 | 17.1 |
| 3−5 days | 324 | 29.3 | 177 | 29.3 | 501 | 29.3 | |
| >5 days | 549 | 49.6 | 369 | 61.0 | 918 | 53.6 | |
| Risk factors | |||||||
| Hypertension | 248 | 22.4 | 135 | 22.3 | 0.966 | 383 | 22.4 |
| Diabetes mellitus | 152 | 13.7 | 73 | 12.1 | 0.330 | 225 | 13.1 |
| Cancer | 66 | 6.0 | 30 | 5.0 | 0.388 | 96 | 5.6 |
| Chronic respiratory disease | 116 | 10.5 | 93 | 15.4 | 0.003 | 209 | 12.2 |
| Chronic heart diseases | 109 | 9.8 | 37 | 6.1 | 0.008 | 146 | 8.5 |
| Obesity | 164 | 14.8 | 112 | 18.5 | 0.047 | 276 | 16.1 |
| Clinical symptoms | |||||||
| Fever | 179 | 16.2 | 126 | 20.8 | 0.016 | 305 | 17.8 |
| Cough | 553 | 50.0 | 400 | 66.1 | <0.001 | 953 | 55.7 |
| Anosmia | 350 | 31.6 | 258 | 42.6 | <0.001 | 608 | 35.5 |
| Ageusia | 354 | 32.0 | 265 | 43.8 | <0.001 | 619 | 36.2 |
| NEWS score | |||||||
| 0−4 | 991 | 89.5 | 502 | 83.0 | <0.001 | 1493 | 87.2 |
| 5−6 | 76 | 6.9 | 43 | 7.1 | 119 | 7.0 | |
| >6 | 40 | 3.6 | 60 | 9.9 | 100 | 5.8 | |
| Clinical outcomes | |||||||
| Death | 11 | 1.0 | 16 | 2.6 | 0.009 | 27 | 1.6 |
| Transfer to intensive care unit | 16 | 1.4 | 31 | 5.1 | <0.001 | 47 | 2.7 |
| Transfer to intensive care unit and/or death |
23 | 2.1 | 44 | 7.3 | <0.001 | 67 | 3.9 |
Dyspnea available within 48 h after admission.
Chi–squared/Fisher's exact test.
The LDCT scan showed signs compatible with pneumonia in 757/1,107 (68.4%) of patients without dyspnea, with 525/757 (69.4%) of patients exhibiting minimal grade 1 lesions, 194/757 (25.6%) of patients exhibiting intermediate grade 2 lesions, and 38/757 (5.0%) of patients exhibiting severe grade 3 lesions (Figure 1 , Table 2 ). Compared to patients without dyspnea, lesions of grades 2 and 3 were significantly more frequent in patients with dyspnea (OR 95% CI: 1.58; 1.24–2.01 and OR 95% CI: 2.15; 1.38−3.37, respectively). One hundred fifty-seven of 1107 (14.2%) patients without dyspnea had an O2 sat ≤95. Among them, 84 had an O2 sat ≤94, 48 had an O2 sat ≤93, 26 had an O2 sat ≤92, 12 had an O2 sat ≤91, and 6 had an O2 sat ≤90. Compared with patients without dyspnea, oxygen saturation was significantly worse in patients with dyspnea (Table 2).
Figure 1.
LDCT scan at admission in a patient with silent COVID-19 pneumonia who abruptly needed O2 support and ICU surveillance for 48 h and his control LDCT at day 10 of treatment with hydroxychloroquine and azithromycin showing residual lesions with retraction, suggesting fibrosis.
Table 2.
Dyspnea, LDCT scan severity scores, and oxygen saturation in patients with COVID-19 pneumonia (n = 1,712).
| No dyspneaa (n = 1107, 65%) |
Dyspneaa (n = 605, 35%) |
Odds ratio 95% confidence interval | |||
|---|---|---|---|---|---|
| n | % | n | % | (Dyspnea vs no dyspnea) | |
| LDCT scanner severity score | |||||
| Normal | 350 | 31.6 | 162 | 26.8 | 0.79 0.64−0.99 |
| Pneumonia | 757 | 68.4 | 443 | 73.2 | 1.27 1.01−1.56 |
| Limited grade 1 | 525 | 47.4 | 248 | 41.0 | 0.77 0.63−0.94 |
| Medium grade 2 | 194 | 17.5 | 152 | 25.1 | 1.58 1.24−2.01 |
| Severe grade 3 | 38 | 3.4 | 43 | 7.1 | 2.15 1.38−3.37 |
| O2 Sat | |||||
| Lower equal 95 | 157 | 14.2 | 121 | 20.0 | 1.51 1.17−1.96 |
| Lower equal 94 | 84 | 7.6 | 72 | 11.9 | 1.65 1.18−2.29 |
| Lower equal 93 | 48 | 4.3 | 44 | 7.3 | 1.73 1.14−2.64 |
| Lower equal 92 | 26 | 2.3 | 30 | 5.0 | 2.17 1.27−3.70 |
| Lower equal 91 | 12 | 1.1 | 19 | 3.1 | 2.96 1.43−6.14 |
| Lower equal 90 | 6 | 0.5 | 15 | 2.5 | 4.67 1.80−12.09 |
Dyspnea available within 48 h of admission and LDCT available within 72 h after admission.
Among the 1,107 without dyspnea, those which LDCT grade 2 and 3 lesions had significantly higher odds of having a poor clinical outcome during follow-up (OR 95% CI: 9.45; 3.02–29.62) (Table 3 ). Compared to the model with O2 sat (OR 95% CI: 0.78; 0.65−0.93), the model with LDCT scan score achieved a better goodness-of-fit (lower AIC).
Table 3.
Factors associated with poor clinical outcome during follow-up (death/transfer to ICUa) among patients without dyspnea – Multivariable logistic regressions (n = 1,107).
| Model A | Model B | Model C | Model D | |
|---|---|---|---|---|
| OR 95%CIc | OR 95%CIc | OR 95%CIc | OR 95%CIc | |
| Sex (ref. Men) | ||||
| Women | 1.40[0.54;3.63] | 1.67[0.62;4.47] | 1.44[0.55;3.80] | 1.65[0.61;4.46] |
| Age (ref. 18−64) | ||||
| >64 | 1.14[0.35;3.69] | 1.11[0.36;3.43] | 1.19[0.37;3.86] | 1.10[0.35;3.42] |
| Risk factors | ||||
| Diabetes mellitus | 0.99[0.34;2.85] | 0.72[0.23;2.23] | 0.88[0.30;2.63] | 0.72[0.23;2.25] |
| Hypertension | 1.90[0.64;5.63] | 1.79[0.60;5.32] | 1.74[0.57;5.28] | 1.65[0.54;5.00] |
| Cancer | 1.21[0.33;4.47] | 1.44[0.40;5.21] | 1.05[0.29;3.90] | 1.30[0.36;4.72] |
| Chronic heart diseases | 3.92[1.34;11.50] | 4.27[1.44;12.66] | 3.83[1.28;11.43] | 4.15[1.37;12.52] |
| Chronic respiratory disease | 0.66[0.16;2.65] | 1.15[0.28;4.70] | 0.70[0.17;2.85] | 1.02[0.25;4.20] |
| Obesity | 1.27[0.39;4.16] | 0.91[0.27;3.12] | 1.11[0.33;3.75] | 0.86[0.24;3.00] |
| Time from symptom onset to admission (ref. < 3days (or no symptom)) | ||||
| Between 3 and 5 days | 1.66[0.46;5.96] | 1.68[0.43;6.47] | 1.44[0.40;5.26] | 1.49[0.39;5.72] |
| >5 days | 1.46[0.45;4.74] | 0.82[0.23;2.92] | 1.32[0.40;4.29] | 0.74[0.21;2.66] |
| NEWS score (ref. 0−4) | ||||
| NEWS >4 | 15.40[4.99;47.49] | 9.64[3.18;29.21] | 9.02[2.72;29.96] | 6.82[2.13;21.87] |
| LDCT scanner severity score (ref. normal/limited) | ||||
| Medium-severe | 9.45[3.02;29.62] | 7.88[2.48;25.00] | ||
| O2 Sat (min = 63–max = 100) | 0.78[0.65;0.93] | 0.82[0.67;1.00] | ||
| AIC/c- statistic | 187 / 0.93 | 172 / 0.93 | 182 / 0.94 | 170 / 0.94 |
| (Area Under ROC Curve) | ||||
a23/1107 (2.1%) of patients were transferred to an ICU/and or died during follow-up.
bList of included risk factors: hypertension, obesity, cancer, diabetes, cardiac disease, and chronic respiratory disease (see Table 1).
Adjusted odds ratio with 95% confidence interval.
In the subset of patients who had undergone at least one blood gas analysis (n = 161) and presented without dyspnea at admission, 28.1% (27/96) presented with a hypoxemia/hypocapnia syndrome defining asymptomatic hypoxia Figure 2 . HCPC analysis distinguished three clusters. Hypoxemia/hypocapnia syndrome (yellow dots) was clustered with death/ICU, elevated NEWS score, age, male, and elevated d-dimers. Hypoxemia/hypocapnia was associated with aging, male, and chronic heart disease but not with diabetes mellitus. Hypoxemia/hypercapnia (33%) clustered with favorable outcome and was associated with younger age. Hyperoxemia/ hypocapnia (22%) was associated with women and younger age. Hyperoxemia/hypercapnia (15%) was surprisingly associated with cancer (Figure 3 , Table 4 ). Hypoxemia/hypocapnia syndrome was strongly associated with death/ICU (OR 95% CI: 4.37; 2.12–9.03) (p < 0.0001) and elevated d-dimers >2.5 mg/l (OR 95% CI: 6.26; 1.99–19.75) (p = 0.002).
Figure 2.
Venn diagram showing that in patients presenting with no dyspnea (27/96) 28.1% will have hypoxemia hypercapnia syndrome defining asymptomatic hypoxia (161 patients).
Figure 3.
Associations between blood gas analysis, clinical data, biological data, and clinical outcomes-Hierarchical Clustering on Principal Components (n = 161).
Table 4.
Clinical characteristics of patients according to hypoxemia/hypocapnia syndrome (n = 161).
| Hypoxemia/ hypocapnia |
Hypoxemia/ hypercapnia |
Hyperxemia/ hypocapnia |
Hyperxemia/ hypercapnia |
All (n = 161) | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| (n = 49, 30%) |
(n = 53, 33%) |
(n = 35, 22%) |
(n = 24, 15%) |
|||||||
| n | % | n | % | n | % | n | % | n | % | |
| Sex | ||||||||||
| Men | 39 | 79.6* | 27 | 50.9 | 13 | 37.1* | 11 | 45.8 | 90 | 55.9 |
| Age at inclusion | ||||||||||
| <45 y.o | 2 | 4.1* | 24 | 45.3* | 12 | 34.3* | 7 | 29.2 | 45 | 28.0 |
| 45−54 y.o | 3 | 6.1 | 5 | 9.4 | 6 | 17.1 | 6 | 25 | 20 | 12.4 |
| 55−64 y.o | 13 | 26.5 | 8 | 15.1 | 11 | 31.4 | 3 | 12.5 | 35 | 21.7 |
| 65−74 y.o | 9 | 18.4 | 5 | 9.4 | 4 | 11.4 | 4 | 16.7 | 22 | 13.7 |
| ≥75 y.o | 22 | 44.9 | 11 | 20.8 | 2 | 5.7 | 4 | 16.7 | 39 | 24.2 |
| Time from symptom onset to admission | ||||||||||
| <3 days (or no symptom) | 11 | 22.5 | 9 | 17.0 | 6 | 17.1 | 8 | 33.3 | 34 | 21.1 |
| 3−5 days | 12 | 24.5 | 17 | 32.1 | 12 | 34.3 | 5 | 20.8 | 46 | 28.6 |
| >5 days | 26 | 53.0 | 27 | 50.9 | 17 | 48.6 | 11 | 45.8 | 81 | 50.3 |
| Risk factors | ||||||||||
| Hypertension | 20 | 40.8 | 14 | 26.4 | 11 | 31.4 | 9 | 37.5 | 54 | 33.5 |
| Diabetes mellitus | 8 | 16.3 | 8 | 15.1 | 2 | 5.7 | 5 | 20.8 | 23 | 14.3 |
| Cancer | 5 | 10.2 | 4 | 7.5 | 0 | 0.0* | 6 | 25.0* | 15 | 9.3 |
| Chronic respiratory disease | 5 | 10.2 | 5 | 9.4 | 3 | 8.6 | 2 | 8.3 | 15 | 9.3 |
| Chronic heart diseases | 16 | 32.7* | 8 | 15.1 | 3 | 8.6 | 4 | 16.7 | 31 | 19.3 |
| Obesity | 6 | 12.2 | 10 | 18.9 | 10 | 28.6 | 9 | 37.5 | 35 | 21.7 |
| Clinical symptoms | ||||||||||
| Fever | 13 | 26.5 | 11 | 20.8 | 10 | 28.6 | 7 | 29.2 | 41 | 25.5 |
| Cough | 28 | 57.1 | 27 | 50.9 | 24 | 68.6 | 14 | 58.3 | 93 | 57.8 |
| Dyspnea | 22 | 44.9 | 19 | 35.8 | 18 | 51.4 | 6 | 25.0 | 65 | 40.4 |
| Anosmia | 8 | 16.3 | 14 | 26.4 | 13 | 37.1 | 5 | 20.8 | 40 | 24.8 |
| Aguesia | 12 | 24.5 | 11 | 20.8 | 13 | 37.1 | 5 | 20.8 | 41 | 25.5 |
| NEWS score | ||||||||||
| 0−4 | 8 | 16.3 | 39 | 73.6 | 22 | 62.9 | 14 | 58.3 | 83 | 51.6 |
| 5−6 | 14 | 28.6 | 6 | 11.3 | 7 | 20.0 | 3 | 12.5 | 30 | 18.6 |
| >6 | 27 | 55.1 | 8 | 15.1 | 6 | 17.1 | 7 | 29.2 | 48 | 29.8 |
| Clinical outcomes | ||||||||||
| Death | 10 | 20.4* | 3 | 5.7 | 2 | 5.7 | 2 | 8.3 | 17 | 10.6 |
| Transfer to intensive care unit | 21 | 42.9* | 6 | 11.3* | 7 | 20.0 | 4 | 16.7 | 38 | 23.6 |
| Transfer to intensive care unit and/or death | 26 | 53.1* | 9 | 17.0* | 8 | 22.9 | 6 | 25 | 49 | 30.4 |
p < 0.05 Fisher’s exact test (versus rest of the sample).
Discussion
In a patient with COVID-19, dyspnea is linked to poor outcome and merits attention and urgent care. However, it is important to underline that among 1712 patients with COVID-19, 64.7% (1107) did not complain of shortness of breath at admission and that 23 of them were transferred to the ICU and/or died highlighting the severity of this clinical presentation. In these patients, News score, LDCT of the thorax, and pulse oximetry are important means to predict death. Moreover, among patients without dyspnea 28.1% presented with hypoxemia/hypocapnia syndrome (happy or silent hypoxemia), which was also strongly associated with a poor outcome. Because of the inclusion criteria, our studied population was significantly more severe and had a longer time from symptom onset to admission, which suggests a selection bias. Consequently, our results cannot be extrapolated to the whole population of patients with COVID-19.
To the best of our knowledge, this work is the most comprehensive comparison of dyspneic and non-dyspneic COVID-19 patients and their blood gas analysis results and lung lesion findings visualized on a low-dose CT scan. Asymptomatic hypoxia has only been reported recently in COVID-19 patients (Xie et al., 2020, Tobin et al., 2020). Hypoxia with accompanying hypocapnia generates no sensation of breathlessness, on the contrary, it may feel comfortable (Ottestad et al., 2020), and COVID-19 patients do not necessarily appear dyspneic until late in the course of the disease (Xie et al., 2020). Physicians treating COVID-19 patients, including us, have reported caring for patients who presented well and then suddenly, within a few hours, progressed to severe respiratory distress (Ottestad et al., 2020).
The subjective sensation of dyspnea is driven by the respiratory center in our brain that is sensitive to increased CO2 and to a lesser degree, decreased O2 (Tobin et al., 2020). Our respiratory center can become desensitized as we age and from chronic conditions such as diabetes (O’Donnell et al., 1988). Patients with diabetes were noted to be 1.8 times more likely to have an impaired ability to perceive respirations than nondiabetic controls. Therefore, it was suggested that patients who present with silent hypoxemia are most likely the elderly and those with chronic diseases and consequently the poorest outcomes. In our cohort, age but not diabetes was associated with dyspnea and hypoxemia. While pulse oximetry underestimates true pO2_mmHg, in patient feeling comfortable at admission, we show a well-predicted outcome. Consequently, in a patient with COVID but without shortness of breath, pulse oximetry should be added to the News score and to LDCT to monitor the care.
The incidence of thrombotic complications in ICU patients with COVID-19 infection is remarkably high (Klok et al., 2020). This has been well established by a recent study of necropsied patients, which reported that although the predominant patterns of lung lesions are diffuse alveolar, damage associated with the presence of platelet-fibrin thrombi in small arterial vessels is consistent with coagulopathy and appears extremely common (Carsana et al., 2020). Hypoxemia and elevated d-dimers strongly suggest that the resulting lung damage is due in part to arterial microemboli and might explain the severity of clinical presentation and the subsequent death. These findings reinforce the recommendation to apply thrombosis prophylaxis in these patients (Klok et al., 2020).
Now, it clearly appears that dyspnea is not a key criterion of initial severity in patients with COVID-19. As recommended by the CDC and many other heath policies throughout the world, “Patients with a mild clinical presentation (absence of viral pneumonia and hypoxia) may not initially require hospitalization, and many patients will be able to manage their illness at home” (Anon, 2020). We believe that it is necessary to systematically ask for shortness of breath that is a subjective sensation, which is not related with respiratory frequency and very frequently reported. Among patients feeling well and without dyspnea, a third (28.1%) might present with hypoxemia at admission. This absence of shortness of breath merits medical attention and it should not be considered as a good sign of well-being. We suggest that for these patients with “a mild clinical presentation” it is particularly important to achieve in a regular basis oxygen saturation with pulse oximetry complete with blood gas analysis if necessary, to allow the early diagnosis of asymptomatic hypoxia and a more appropriate care to reduce the poor outcome
Conflict of interest
None declared.
Financial support
This study was supported by the National Research Agency under the program «Investissements d’avenir», reference ANR-10-IAHU-03.
Ethical approval
The noninterventional retrospective nature of the study was approved by our institutional review board committee (Mediterranée Infection No.: 2020-021). According to European General Data Protection Regulation No 2016/679, patients were informed of the potential use of their medical data and that they could refuse. The analysis of collected data followed the reference methodology MR-004 registered on No. MR 5010010520 in the AP-HM register.
Acknowledgments
The English language in this text was edited by American Journal Experts.
References
- Anon . 2020. Interim Clinical Guidance for Management of Patients with Confirmed Coronavirus Disease (COVID-19): Clinical Managment and Treatment.https://wwwcdcgov/coronavirus/2019-ncov/hcp/clinical-guidance-management-patientshtml#clinical-management-treatment%3C [Google Scholar]
- Boudjema S., Finance J., Coulibaly F., Meddeb L., Tissot-Dupont H., Michel M. Olfactory and gustative disorders for the diagnosis of COVID-19. Travel Med Infect Dis. 2020;(September):101875. doi: 10.1016/j.tmaid.2020.101875. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Carsana L., Sonzogni A., Nasr A., Rossi R.S., Pellegrinelli A., Zerbi P. Pulmonary post-mortem findings in a series of COVID-19 cases from northern Italy: a two-centre descriptive study. Lancet Infect Dis. 2020;(June) doi: 10.1016/S1473-3099(20)30434-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Couzin-Frankel J. The mystery of the pandemic’s ‘happy hypoxia’. Science. 2020;368(May (6490)):455–456. doi: 10.1126/science.368.6490.455. [DOI] [PubMed] [Google Scholar]
- Guan W., Xian J. The progress of 2019 Novel Coronavirus (2019-nCoV) event in China. J Med Virol. 2020;(February) doi: 10.1002/jmv.25705. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Guan W.J., Ni Z.Y., Hu Y., Liang W.H., Ou C.Q., He J.X. Clinical characteristics of coronavirus disease 2019 in China. N Engl J Med. 2020;382(April (18)):1708–1720. doi: 10.1056/NEJMoa2002032. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Huang C., Wang Y., Li X., Ren L., Zhao J., Hu Y. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. Lancet. 2020;395(February (10223)):497–506. doi: 10.1016/S0140-6736(20)30183-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Klok F.A., Kruip M.J.H.A., van der Meer N.J.M., Arbous M.S., Gommers D., Kant K.M. Confirmation of the high cumulative incidence of thrombotic complications in critically ill ICU patients with COVID-19: an updated analysis. Thromb Res. 2020;191(July):148–150. doi: 10.1016/j.thromres.2020.04.041. [DOI] [PMC free article] [PubMed] [Google Scholar]
- La S.B., Le B.M., Andreani J., Hoang V.T., Grimaldier C., Colson P. Viral RNA load as determined by cell culture as a management tool for discharge of SARS-CoV-2 patients from infectious disease wards. Eur J Clin Microbiol Infect Dis. 2020;39(June (6)):1059–1061. doi: 10.1007/s10096-020-03913-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lagier J.C., Million M., Gautret P., Colson P., Cortaredona S., Giraud-Gatineau A. Outcomes of 3,737 COVID-19 patients treated with hydroxychloroquine/azithromycin and other regimens in Marseille, France: a retrospective analysis. Travel Med Infect Dis. 2020 doi: 10.1016/j.tmaid.2020.101791. In press. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lê S., Josse J., usson F. FactoMineR: an R package for multivariate analysis. J Stat Softw. 2008;25(1):1–18. [Google Scholar]
- Liao X., Wang B., Kang Y. Novel coronavirus infection during the 2019-2020 epidemic: preparing intensive care units-the experience in Sichuan Province, China. Intensive Care Med. 2020;46(February (2)):357–360. doi: 10.1007/s00134-020-05954-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- O’Donnell C.R., Friedman L.S., Russomanno J.H., Rose R.M. Diminished perception of inspiratory-resistive loads in insulin-dependent diabetics. N Engl J Med. 1988;319(November (21)):1369–1373. doi: 10.1056/NEJM198811243192102. [DOI] [PubMed] [Google Scholar]
- Ottestad W., Seim M., Maehlen J.O. COVID-19 with silent hypoxemia. Tidsskr Nor Laegeforen. 2020;140(May (7)) doi: 10.4045/tidsskr.20.0299. [DOI] [PubMed] [Google Scholar]
- Tobin M.J., Laghi F., Jubran A. Why COVID-19 silent hypoxemia is baffling to physicians. Am J Respir Crit Care Med. 2020;202(August (3)):356–360. doi: 10.1164/rccm.202006-2157CP. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Xie J., Tong Z., Guan X., Du B., Qiu H. Clinical characteristics of patients who died of coronavirus disease 2019 in China. JAMA Netw Open. 2020;3(April (4)) doi: 10.1001/jamanetworkopen.2020.5619. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zhou F., Yu T., Du R., Fan G., Liu Y., Liu Z. Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study. Lancet. 2020;(March) doi: 10.1016/S0140-6736(20)30566-3. [DOI] [PMC free article] [PubMed] [Google Scholar]