47 patients were screened and 27 were finally recruited. 15 (55.5%) of them required intubation and IMV. 73.3% of intubated patients were male and 3 patients (20%) had diabetes as a previous diagnosis. Arterial hypertension was more common among intubated patients compared to non-intubated (60% vs 16.7%, p 0.047) and renin-angiotensin-aldosterone system inhibitors (RAASi) were the more common treatments (53.3% vs 8.3%, p 0.019). Interestingly, betablockers were not commonly used in these patients (6.7% in intubated patients).
Patients who needed to be intubated and mechanically ventilated had a more rapid evolution to disease as they were admitted to the hospital more precociously (5 days vs 8 days, p 0.029) and were sicker (APACHE II score 14 vs 8.5, p 0.037), although the stay in the general ward previously to the ICU admission was the same (2 days, p 0.856).
The median from ICU admission to intubation was 5.5 days (IQR 2.75-8) and IMV duration 17 days (7–35).
No patients died in the non-intubated group however mortality at 28 days was 40% in the intubated group (p 0.02). Also, 53.3% of the intubated patients needed renal replacement therapies versus none in the non-intubated group (p 0.003). As expected, hospital and ICU length-of-stay were longer in the intubated group (37 vs 17 days and 33 vs 10 days, p < 0.001; respectively).
Table 1
Demographics, previous treatments and clinical results. RAASi, renal angiotensin-aldosterone system inhibitors, RRT renal replacement therapies; LOS length of stay
| Non-Intubated (n = 12) | Intubated (n = 15) | p |
Male sex, n (%) | 8 (66.7) | 11 (73.3) | 1 |
Age, median (IQR) | 51 (28–68) | 69 (61–74) | 0.021 |
Body mass index, median (IQR) | 28 (25.125–33.8) | 26.1 (24.4–29.4) | 0.399 |
APACHE score, median (IQR) | 8.5 (5-12.25) | 14 (9–16) | 0.037 |
Diabetes, n (%) | 4 (33.3) | 3 (20) | 0.662 |
Arterial hypertension, n (%) | 2 (16.7) | 9 (60) | 0.047 |
Betablockers, n (%) | 2 (16.7) | 1 (6,7) | 0.569 |
RAASi, n (%) | 1 (8.3) | 8 (53.3) | 0.019 |
Days from symptoms to hospital admission, median (IQR) | 8 (6–9) | 5 (2–7) | 0.029 |
Days from hospital admission to ICU, median (IQR) | 2 (0–3) | 2 (1–3) | 0.856 |
Antibiotics at day 1, n (%) | 4 (33.3) | 3 (20) | 0.662 |
Mortality 28 days, n (%) | 0 | 6 (40) | 0.02 |
Need of RRT, n (%) | 0 | 8 (53.3) | 0.003 |
LOS ICU, median (IQR) | 10 (5–12) | 33 (22–49) | < 0.001 |
LOS Hospital, median (IQR) | 17 (10.75-18) | 37 (24–49) | < 0.001 |
Time to intubation, median (IQR) | | 5.5 (2.75-8) | |
Days duration mechanical ventilation, median (IQR) | | 17 (7–35) | |
CLINICAL AND BIOCHEMICAL PARAMETERS (Table 2):
Intubated patients were slightly more tachypneic than those non-intubated, although non significantly (26 vs 23.5 breath/min, p 0.2). Patients who required intubation were more hypoxic as described by the ROX index (5.14 vs 7.23, p 0.013) and SAF index (122 vs 154, p 0.031), requiring higher Fi02 (80 vs 60%, p 0.037).
Interestingly, cardiac biomarkers, troponin I and ProBNP were also higher (0.11 vs 0.01 and 361 vs 114 pg/mL, respectively), the last one almost reaching significance (p 0.167 and 0.068).
Table 2
Clinical and biochemical parameters. RR respiratory rate, HFO high Flow oxygen therapy, Fi02% inspired oxygen fraction, LDH lactate deshidrogenase, ProBNP Pro Brain natriuretic peptide, ProADM pro-Adrenomedullin, CRP C-reactive protein, IL-6 interleukin 6.
| Non-Intubated (n = 12) | Intubated (n = 15) | p |
RR breaths/min, median (IQR) | 23.5 (20.25-26) | 26 (20–31) | 0.2 |
HFO flow L/min, median (IQR) | 50 (50–60) | 60 (50–70) | 0.083 |
Fi02%, median (IQR) | 60 (52.5–85) | 80 (70–100) | 0.037 |
ROX index, median (IQR) | 7.23 (5-9.33) | 5.14 (3.25–6.05) | 0.013 |
SAFI, median (IQR) | 154 (141-186.25) | 122 (96–138) | 0.031 |
Plasma creatinine mg/dL, median (IQR) | 0.72 (0.55–0.89) | 0.63 (0.55–0.84) | 0.683 |
Plasma urea mg/dL, median (IQR) | 37.5 (29.5–52.5) | 50 (38–64) | 0.152 |
LDH mg/dL, median (IQR) | 844.5 (538–992) | 959 (810–1074) | 0.152 |
Troponin I ng/L, median (IQR) | 0.01 (0.007–0.115) | 0.11 (0.1–0.2) | 0.167 |
ProBNP pg/mL, median (IQR) | 114 (25–268) | 316 (151–393) | 0.068 |
CRP mg/L, median (IQR) | 7.89 (2.2–124) | 11.6 (4.55–31.4) | 0.981 |
Procalcitonin ng/mL, median (IQR) | 0.11 (0.1-0.175) | 0.15 (0.1–0.6) | 0.152 |
IL-6 pg/dL, median (IQR) | 19.56 (11.72–24.17) | 28.75 (17–52) | 0.129 |
D-dimer ng/mL, median (IQR) | 1029 (615-1155.55) | 958.5 (660.5–2155) | 0.829 |
Ferritin mg/dL, median (IQR) | 1011 (579–1491) | 1192 (751–2407) | 0.531 |
ProADM nmol/L, median (IQR) | 0.8 (0.63–0.93) | 0.8 (0.73–1.33) | 0.403 |
Leucocytes, median (IQR) | 9600 (6050–10825) | 9000 (7600–11100) | 0.943 |
HRV ANALYSIS (Table 3):
Recording lasted a median of 38 minutes in intubated patients vs 43 minutes in those who did not need to be intubated. Heart rate was 80 beats/min in each group. Median RR interval was also similar.
HRV, defined by time domain metrics, was similar between groups (SDNN 27.15 vs 29.21 msec, p 0.829; pNN50 0.21 vs 0.16, p 0.905 and rMSSD 14.7 vs 11.14, p 1).
There were no statistical differences between intubated and non-intubated patients in frequency domain analysis: total power was 515.29 vs 502.33 msec2; normalized LF was lower (58 vs 78%) and normalized HF higher (42 vs 21%). Therefore, LF/HF ratio was lower in intubated patients (1.38 vs 3.65) but not reaching statistical significance (p 0.167).
We found no differences using nonlinear dynamics analysis. SD1 and SD2 derived from Poincaré plots were similar between groups (7.9 vs 10.24 and 40.66 vs 37.03, respectively). DFA alpha1 was slightly lower in intubated patients (1.08 vs 1.11) but alpha2 was 1.2 in each group. Sample entropy was also similar (0.31 vs 0.33).
When comparing with healthy controls, in sitting position like our patients, taken from previous studies [42, 43] we found that, although lnLF/HF was not different, all others metrics analyzed, including heart rate, were lower in Covid-19 patients than in healthy individuals. Mean RR was also lower in Covid-19, in intubated patients almost reaching statistical significance, p 0.06 (Table 1, Supplemental material).
Table 3
HRV metrics. Values are expressed as median and interquartile range. DFA detrended fluctuation analysis.
| Non-intubated (n = 12) | Intubated (n = 15) | p |
Record duration, min | 43 (17–46) | 38 (35–43) | 0.581 |
Heart rate, beats/min | 80 (66–106) | 80 (69–85) | 0.829 |
RR interval, msec | 749 (565–903) | 752 (709–867) | 0.755 |
SDNN, msec | 27.15 (19.25–41.46) | 29.21 (21.61–51.5) | 0.829 |
SDANN, msec | 15.94 (10.32–28.69) | 18.06 (10.4-30.88) | 0.838 |
pNN50, % | 0.21 (0-1.43) | 0.16 (0-9.66) | 0.905 |
SDSD, msec | 14.71 (8.13–23.68) | 11.15 (9.41–29.77) | 1 |
IRRR, msec | 40 (28–64) | 40 (28–72) | 0.867 |
rMSSD, msec | 14.7 (8.13–23.68) | 11.14 (9.4-29.77) | 1 |
HRV index, msec | 8.6 (6.17–13.63) | 8.7 (5.79–15.2) | 0.829 |
Total power, msec2 | 502.33 (294.5-1159.03) | 515.29 (322.42-1398.43) | 0.792 |
LFn, % | 78 (53–89) | 58 (47–79) | 0.167 |
HFn, % | 21 (11–47) | 42 (21–53) | 0.167 |
LF/HF | 3.65 (1.15–8.13) | 1.38 (0.9–3.68) | 0.167 |
SD1 | 10.42 (5.77–16.75) | 7.9 (6.69–21.06) | 0.981 |
SD2 | 37.03 (24.13–57.98) | 40.66 (29.34–66.73) | 0.719 |
SD1/SD2 | 0.26 (0.14–0.34) | 0.28 (0.18–0.35) | 0.548 |
DFA α1 | 1.11 (0.86–1.51) | 1.08 (0.78–1.2) | 0.456 |
DFA α2 | 1.2 (1.12–1.27) | 1.2 (0.98–1.3) | 0.683 |
Sample entropy | 0.33 (0.22–0.35) | 0.31 (0.17–0.51) | 0.829 |