Characteristics and outcomes of patients with COVID-19 in intensive care in the first year of the pandemic: A Colombian observational study

Abstract

Background

The COVID-19 pandemic has had a devastating impact on the world, mainly during the first year of the pandemic, where strategies such as vaccination were not available. Information on the outcomes of patients admitted to the intensive care unit (ICU) in Colombia is scarce. Our main objective was to characterize critically ill patients with COVID-19 in our region.

Methods

We conducted a single-center retrospective observational study in which we included patients with COVID-19 confirmed by RT-PCR who were admitted to the adult ICU between March 18, 2020 and March 18, 2021, in Quindío, Colombia. We identify the clinical and laboratory characteristics at admission, the support used, and their relationship with mortality during ICU hospitalization.

Results

Three hundred and fifty-nine patients with a confirmed diagnosis of COVID-19 were admitted, 64% men, mean age was 62.7 years (SD ± 12.3), body mass index 27.9 kg/m² (±5.8), SOFA score was 7.6 (±3.12), Pa/FiO₂ 96.2 (±62.3), and lung compliance 30.5 ml/cmH₂O (±18.4). Mortality was 60%. The variables with the highest mortality association were obesity OR: 2.38 (95% CI: 1.39–4.09, p: <.001), Glasgow coma scale at admission <12: 17.5, (5.21–58.8, p: <.001), PaFiO₂ <100: 5.63, (3.38–9.39, p: <.001), static lung compliance less than 50 ml/cmH₂O: 3.54, (3.38–9.39, p: <.001), SOFA score >5: 3.75 (2.19–6.42, p: <0.001), ferritin > 1000: 2.58, (1.66–4.02, p: <.001), C-reactive protein > 5: 2.52 (1.42–4.26, p: <.001), and LDH > 280: 2.71 (1.55–4.74, p: <.001). Patients who required PEEP > 10 cmH₂O: 2.34 (1.48–3.70, p: <.001), FiO₂ > 60%: 4.01, (2.46–6.53, p: <.001), and ventilation in the prone position.

Conclusion

Mortality in the first year of the pandemic in our region was high, mainly associated with obesity, inflammation, altered mental status upon admission, and increased lung elastance.

Resumen

Contexto

La pandemia de COVID-19 ha tenido un impacto devastador en el mundo, principalmente durante el primer año de la pandemia, donde no se disponía de estrategias como la vacunación. La información sobre los resultados de los pacientes ingresados en la unidad de cuidados intensivos (UCI) en Colombia es escasa. Nuestro principal objetivo fue caracterizar a los pacientes críticos con COVID-19 en nuestra región.

Métodos

Realizamos un estudio observacional retrospectivo unicéntrico donde incluimos pacientes confirmados con RT-PCR para COVID-19 que ingresaron a la UCI de adultos entre el 18 de marzo de 2020 y el 18 de marzo de 2021 en Quindío, Colombia. Identificamos las características clínicas y de laboratorio al ingreso, los soportes utilizados y su relación con la mortalidad durante la hospitalización en UCI.

Resultados

Ingresaron 359 pacientes con diagnóstico confirmado de COVID-19, 64% hombres, edad 62,7 años (DE ± 12,3), índice de masa corporal 27,9 kg/m² (± 5,8), SOFA score 7,6 (± 3,12), Pa/FiO₂ 96,2 (± 62,3) y distensibilidad pulmonar 30,5 ml/cmH₂O (± 18,4). La mortalidad fue del 60%. Las variables con mayor asociación a mortalidad fueron la obesidad, OR: 2,38 (IC 95%: 1,39-4,09, p < 0,001), escala de coma de Glasgow al ingreso < 12: 17,5 (5,21-58,8, p < 0,001), PaFiO₂ < 100: 5,63 (3,38-9,39, p < 0,001), distensibilidad pulmonar estática inferior a 50 ml/cmH₂O: 3,54 (3,38-9,39, p < 0,001), puntuación SOFA > 5: 3,75 (2,19-6,42, p < 0,001), ferritina > 1.000: 2,58 (1,66-4,02, p < 0,001), proteína C reactiva > 5: 2,52 (1,42-4,26, p < 0,001) y LDH > 280: 2,71 (1,55-4,74, p < 0,001). Pacientes que requirieron PEEP > 10 cmH₂O: 2,34 (1,48-3,70, p < 0,001), FiO₂ > 60%: 4,01 (2,46-6,53, p < 0,001) y ventilación en decúbito prono.

Conclusión

La mortalidad en el primer año de la pandemia en nuestra región fue alta, asociada principalmente a obesidad, inflamación, alteración del estado mental al ingreso y aumento de la elastancia pulmonar.

Introduction

Emerging infections are a real challenge for health personnel. Currently, SARS-CoV-2 has shown an infectious dynamic that still has the world on alert after more than a year and a half of the pandemic,^1,2 with more than 560 million infected patients and more than 6 million deaths, up to July 15, 2022.³

There are 1119 municipalities with COVID-19 cases in Colombia, which corresponds to 99% of the territory. As of September 21, 2021, 4.9 million people have been infected, and about 126,000 patients have died from COVID-19.^4,5 To face this crisis, the country has increased its installed beds, currently having 46,368 hospital beds, 3795 intermediate care beds and 11,450 intensive care beds.^6,7

Quindío, a small state located in the center of Colombia, has approximately 500,000 inhabitants and has 563 available hospital beds, 53 intermediate care beds and 150 ICU beds. 85% (480) of hospital beds are concentrated in the capital city (Armenia), 100% of unique care beds. Total number of ICU beds represent 300% growth in installed intensive care capacity since the onset of the pandemic.⁷

In the literature, some works account for the clinical and laboratory characteristics of patients infected with SARS-COV-2 in Asian populations,⁸ European^9–11 and American populations.¹² For example, 20% of patients require admission to the ICU due to critical illness; cardiovascular disease, obesity, and diabetes are risk factors for severe disease and unfavorable outcomes.¹³ In addition, 17% of patients require mechanical ventilation, with men more likely than women to receive ventilatory support (22% versus 12%). Ventilated patients have a mean duration of mechanical ventilation of 12 days and 18 days of hospitalization. The mortality of ventilated patients was 53%.¹⁰

In Latin America, the information scarce.^14–16 For example, not much information is available in Colombia, where most of the population is different from the Asian, European, and Anglo-Saxon. The existing one is from studies of short duration or with small samples,¹⁷ which leaves us with a knowledge gap at the national level regarding the characteristics, management, and outcomes of critically ill patients due to COVID-19 who are admitted to the ICU as adults.

This study aimed to identify the baseline characteristics, treatments, and outcomes of patients infected with SARS-COV-2 in critical condition who were hospitalized in the ICU of the San Juan de Dios regional hospital in the first year of the pandemic.

Methods

Study design

This is an observational, retrospective, single-center study.

Settings

It was carried out in the intensive care unit of the regional hospital of Quindío San Juan de Dios; this is the leading and most prominent university health center in the state, it currently has 304 beds, of which 18 are intensive care for adult patients with COVID-19 and 56 unique care beds, built progressively since the beginning of the pandemic as an ICU extension.

The ICU is attended 24 h a day by two intensive care physicians, two nursing professionals, two respiratory therapy professionals, six nursing assistants, a speech therapist, a physiotherapist, and a nutritionist.

The data was analyzed for the corresponding period between March 18, 2020, and March 18, 2021.

Participants

Describe the data related to patients with a confirmed diagnosis of COVID-19 by RT-PCR in any of its manifestations of the spectrum of the disease who have been admitted to the ICU as adults during the period studied.

Variables

Anthropometric variables, comorbidities, admission manifestations, disease severity are described using the SOFA score, ventilatory strategies, pulmonary mechanics, use of prone position, oxygenation disorder (severity), the phenotype of pulmonary manifestations (L–H), internal environment variables (renal function, electrolytes), inflammation variables (C-reactive protein, lactic dehydrogenase, ferritin). Coagulation variables (platelets – D-dimer), myocardial injury variables (troponin I), as well as the use of hemodynamic support drugs, antibiotics, sedatives, analgesics and neuromuscular relaxants, anticoagulant and anti-inflammatory drugs, outcome variables (mortality).

Source of information

A database was prospectively completed in a spreadsheet (Excel®) from patient one to the cut-off date of the study.

Bias control

The database has all the variables in analysis; It has no loss of information; it was completed in real-time by an intensivist (CASP) and verified by two intensivists (WCO-YDG).

Statistical analysis

Statistical analysis was performed in SPSS 27® (IBM, USA). The qualitative variables are summarized with proportions and the quantitative variables with measures of central tendency and dispersion. Normality was assessed with the Kolmogorov–Smirnov test.

Association measures (Odds Ratio) and χ² were performed to compare the categorical variables. Statistical significance will be expressed as a function of p. <0.05.

Ethical aspects

The bioethics committee approved this study of the San Juan de Dios University Hospital, Armenia, Quindío, Colombia, with registration No.: 02461/21.

Results

In total, 359 patients with a confirmed diagnosis of COVID-19 were admitted, 232 men (64%), the mean age was 62.7 years (SD: ±12.3), the body mass index (BMI) was 27.9 kg/m² (SD: ±5.8), the mean of the SOFA score (Sequential Organ Failure Assessment) was 7.6 (SD: ±3.12), the mean of Pa/FiO₂ and lung compliance at admission were low (96.2, SD: ±62.3) and (30.5 ml/cmH₂O, SD: ±18.4) respectively. On the other hand, inflammation variables were high at admission: Ferritin (1312 μg/L, DS: ±1251.2), C-reactive protein (14.4 mg/L, DS: ±10), lactic dehydrogenase (469.6 U/L, DS: ±185.3) as shown in Table 1.

Table 1.

Quantitative variables of the patients at admission.

Variable	Mean	SD
Age, y	62.7	12.3
BMI, kg/m^2	27.9	5.08
SOFA	7.6	3.12
MAP, mmHg	89.1	18.09
ICU's Hosp, d	8.5	6.22
Mechanical ventilation, d	7.1	5.91
FiO2, %	74.7	21.60
PEEP, cmH2O	9.1	3.33
Plateau, cmH2O	22.6	8.74
Compliance, ml/cmH2O	30.5	18.47
Driving pressure, cmH2O	13.4	6.52
Prone position, hours	11.4	8.69
Prone position, d	1.9	1.45
P/F income	96.2	62.37
Hemoglobin, g/L	12.4	2.30
Platelets, per 10^9/μL	245.5	113.48
White blood cells, per 10^9/μL	14.7	7.6
Neutrophil, per 10^9/μL	8	10
Lymphocyte, per 10^9/μL	0.7	0.7
D-dimer, mg/L	690.8	1249.52
Ferritin, μg/L	1312.3	1251.28
C-reactive protein, mg/L	14.4	10.08
LDH, U/L	469.6	185.34
Creatinine, mg/dL	1.7	2.08
Troponin, ng/L	0.4	1.63
Na	138.8	5.47
K	4.6	0.83
Ca	7.4	4.47
Cl	104.6	9.32

The most frequent symptoms were cough (58.8%) and respiratory distress (83.8%), hypertension was the most prevalent morbidity (52.4%), most of the patients required mechanical ventilation (90.3%), and in 88.3% of the cases, the assisted-controlled ventilation mode was used.

88.3% of the patients required sedation and 66.3% neuromuscular relaxation; 97.2% received anticoagulants, 86.4% corticosteroids, 94.2% antibiotics, and less than a third of the patients required vasoactive drugs (28.7%). Unfortunately, 60% of patients admitted to the adult ICU for COVID-19 died (Tables 2 and 3).

Table 2.

Qualitative characteristics at admission.

Variable	n	%
Sex
Women	127	35.4
Men	232	64.6
Outcome
Alive	142	39.6
Death	217	60.4
Morbidities
None	102	28.4
Hypertension	188	52.4
Type 2 diabetes	93	25.9
Obesity	88	24.5
Symptoms
Cough
No	148	41.2
Yes	211	58.8
Fever
No	169	47.1
Yes	190	52.9
Respiratory difficult
No	58	16.2
Yes	301	83.8
Diarrhea
No	296	82.5
Yes	63	17.5
Headache
No	263	73.3
Yes	96	26.7
Olfactory disorder
No	325	90.5
Yes	34	9.5
Neurological focalization
No	335	93.3
Yes	24	6.7

Statistically significant results are highlighted in bold.

Table 3.

Treatments used for the patients.

Treatment	n	%
Mechanical ventilatory mode
None	35	9.7
ACV	317	88.3
ACP	5	1.4
PRVC	1	0.3
Duolevel	1	0.3
Sedation and analgesia
Fentanyl	317	88.3
Midazolam	224	62
Propofol	5	1.4
Dexmedetomidine	6	1.7
Muscular relaxant
None	121	33.7
Rocuronium	230	64.1
Cisatracurio	8	2.2
Anticoagulation/prophylaxis
No	10	2.8
Yes	349	97.2
Corticoids
No	49	13.6
Yes	310	86.4
Antibiotics
None	21	5.8
Meropenem	44	12.3
Pip/tazobactam	130	36.2
Cefepime	18	5
Ampicillin/sulbactam	113	31.5
Vancomycin	33	9.2
Macrolides	158	44
RBC transfusion
No	352	98.1
Yes	7	1.9
Nutritional support
None	32	8.9
Enteral	326	90.8
Parenteral	1	0.3
Vasoactive/inotropic
None	256	71.3
Norepinephrine	98	27.3
Vasopressin	3	0.8
Dobutamine	2	0.6
Renal support
No	317	88.3
Yes	42	11.7

Mortality in men was slighter (61.6% vs. 58.3%), lower in patients under 50 years of age (46.4%), and with a normal body mass index (50.5%). Mortality was higher when admission symptoms were cough (64.9%), headache (63.3%), and respiratory distress (62.8%) (Table 4).

Table 4.

Variables and mortality.

Characteristics	Outcome		% Mortality
	Alive	Dead
Sex
Men	89	143	61.6
Women	53	74	58.3
Age, y
<50	30	26	46.4
50–60	31	47	60.3
60–70	51	78	60.5
>70	30	66	68.8
BMI, kg/m^2
<25	55	56	50.5
25–30	56	98	63.6
>30	31	63	67.0
Symptoms
Cough
No	68	80	54.1
Yes	74	137	64.9
Fever
No	69	100	59.2
Yes	73	116	61.4
Headache
No	109	154	58.6
Yes	32	63	66.3
Diarrhea
No	118	178	60.1
Yes	24	39	61.9
Respiratory difficult
No	30	28	48.3
Yes	112	189	62.8
Neurological focalization
No	125	210	62.7
Yes	17	7	29.2
Olfactory disorders
No	128	197	60.6
Yes	14	20	58.8
Symptoms – ICU, d
<5	48	52	52.0
5–8	44	66	60.0
8–11	31	59	65.6
>11	19	40	67.8
Morbidities
HTA
No	70	101	59.1
Yes	72	116	61.7
Type 2 DM
No	112	154	57.9
Yes	30	63	67.7
Measurements on admission
P/F
<100	76	188	71.2
100–200	44	23	34.3
200–300	18	6	25.0
>300	4	0	0.0
GCS
<8	109	205	65.3
8–12	5	9	64.3
>12	28	3	9.7
MAP, mmHg
<65	12	20	62.5
65–90	67	111	62.4
>90	63	86	57.7
SOFA
<5	48	26	35.1
5–10	77	135	63.7
>10	17	56	76.7
D-dimer
<500	80	104	56.5
>500	62	113	64.6
C-reactive protein
<5	42	31	42.5
5–15	53	85	61.6
>15	47	101	68.2
Ferritin
<300	27	21	43.8
300–1000	69	76	52.4
1000–1700	25	39	60.9
>1700	21	81	79.4
	119	170	58.8
>Pc99	23	47	67.1
Troponin
Creatinine
<1.35	107	140	56.7
>1.35	35	77	68.8
LDH
<280	37	25	40.3
>280	105	192	64.6
Compliance phenotipe
L	42	23	35.9
H	100	194	66.0
Ventilatory support
x PEEP
<5	32	5	13.5
5–10	72	112	60.9
>10	38	100	72.5
Prone position
No	68	64	48.5
Yes	74	153	67.4
VM days
<5	63	97	60.6
5–10	42	78	65.0
10–15	17	32	65.3
>15	20	10	33.3
Other support
Vasoactive and inotropic
No	111	145	56.6
Yes	31	72	69.9
Thrombo-prophylaxis
No	7	3	30.0
Yes	135	214	61.3
RBC transfusion
No	138	214	60.8
Yes	4	3	42.9
Renal support
No	124	193	60.9
Yes	18	24	57.1
ICU's Hosp, d
<5	39	93	70.5
5–10	45	78	63.4
10–15	27	31	53.4
>15	31	15	32.6

Statistically significant results are highlighted in bold.

The variables with the highest mortality association were age over 50 years (OR: 1.97, 95% CI: 1.11–3.50, p: 0.020), presence of respiratory distress (OR: 1.81, 95% CI: 1.03–3.18, p: 0.038), obesity was the morbidity mostly associated with mortality (OR: 2.38, 95% CI: 1.39–4.09, p: <0.001). The measurements at admission that are most associated with mortality were the Glasgow coma scale <12 (OR: 17.5, 95% CI: 5.21–58.8, p: <0.001), PaFiO₂ < 100 (OR: 5.63, 95% CI: 3.38–9.39, p: <0.001), static lung compliance less than 50 ml/cmH₂O (OR: 3.54, 95% CI: 3.38–9.39, p: <0.001), SOFA score > 5 (OR: 3.75, 95% CI: 2.19–6.42, p: <0.001), ferritin > 1000 (OR: 2.58, 95% CI: 1.66–4.02, p: <0.001), C-reactive protein > 5 (OR: 2.52, 95% CI: 1.42–4.26, p: <0.001), and LDH > 280 (OR: 2.71, 95% CI: 1.55–4.74, p: <0.001).

Regarding respiratory supports, patients who required PEEP > 10 cmH₂O (OR: 2.34, 95% CI: 1.48–3.70, p: <0.001), FiO₂ > 60% (OR: 4.01, 95% CI: 2.46–6.53, p: <0.001) and ventilation in the prone position had a greater association with mortality (Table 5).

Table 5.

Association between variables and mortality.

Characteristics	Outcome		OR	95% CI		P value
	Dead (n:217)	Alive (n:142)
Demographic variables
Sex
Women	74	53	0.87	0.56	1.35	0.53
Men	143	89
Age (y)
>50	191	112	1.97	1.11	3.50	0.020
<50	26	30
Symptoms
Cough
Yes	137	74	1.57	1.02	2.42	0.038
No	80	68
Respiratory difficult
Yes	189	112	1.81	1.03	3.18	0.038
No	28	30
Fever
Yes	116	73	1.10	0.72	1.68	0.67
No	100	69
Headache
Yes	63	32	1.39	0.85	2.28	0.18
No	154	109
Diarrhea
Yes	39	24	1.08	0.62	1.88	0.79
No	178	118
Neurological focalization
Yes	7	17	0.25	0.10	0.61	0.001
No	210	125
Olfactory disorders
Yes	20	14	0.93	0.45	1.90	0.83
No	197	128
Symptoms to ICU (d)
>5	165	94	1.62	1.02	2.59	0.04
<5	52	48
Morbidities
Hypertension
Yes	116	72	1.12	0.73	1.71	0.61
No	101	70
Type 2 diabetes
Yes	63	30	1.53	0.93	2.51	0.09
No	154	112
Obesity
Yes	66	22	2.38	1.39	4.09	0.001
No	151	120
Measurements on admission
Glasgow Coma Scale
<12	214	114	17.5	5.21	58.8	<0.001
>12	3	28
MAP (mmHg)
<65	20	12	1.10	0.52	2.33	0.80
>65	197	130
Pa/FiO2
<100	188	76	5.63	3.38	9.39	<0.001
>100	29	66
Compliance (ml/cmH2O)
<50	194	100	3.54	2.02	6.22	<0.001
>50	23	42
SOFA
>5	191	94	3.75	2.19	6.42	<0.001
<5	26	48
D-Dimer
>500	113	62	1.4	0.94	2.15	0.11
<500	104	80
Ferritin
>1000	120	46	2.58	1.66	4.02	<0.001
<1000	97	96
C-reactive protein
>5	186	100	2.52	1.49	4.26	<0.001
<5	31	42
LDH
>280	192	105	2.71	1.55	4.74	<0.001
<280	25	37
Creatinine
>1.35	77	35	1.68	1.05	2.70	0.03
<1.35	140	107
Troponin I
>P.99	47	23	1.43	0.82	2.48	0.20
	170	119
Ventilatory support on admission
PEEP
>10	100	38	2.34	1.48	3.70	<0.001
<10	117	104
FiO2(%)
>60	181	79	4.01	2.46	6.53	<0.001
<60	36	63
Prone position
No	64	68	0.46	0.29	0.71	<0.001
Yes	153	74	2.20	1.42	3.41
Other support
Vasoactive and inotropic
Yes	72	31	1.78	1.09	2.9	0.03
No	145	111
Thromboprophylaxis
Yes	214	135	3.70	0.94	14.55	0.04
No	3	7
RBC transfusion
Yes	3	4	0.48	0.11	2.19	0.33
No	214	138
Renal support
Yes	24	18	0.86	0.45	1.64	0.64
No	193	124
Corticoids
Yes	207	103	7.84	3.76	16.33	<0.001
No	10	39

Statistically significant results are highlighted in bold.

Discussion

To our knowledge, this is the largest study on the characteristics and outcomes of adult COVID-19 patients admitted to the ICU in Colombia, a lower-middle-income Latin American country. In this study, most patients were men over 60 years of age, hypertensive and overweight; it was like that found in other Latin American studies. In a Mexican study, the average age of patients admitted to the ICU was 57 years, the body mass index (BMI) was 30.7 kg/m² (± 5.5), and 63% of the patients had a history of hypertension,¹⁵ and in an Argentine study, the median was 62 years, the BMI was 29 kg/m²,^26–31 and 47% of patients were hypertensive.¹⁶ In a Colombian study it was found that in seriously ill patients the age, on average, was 59.3 years, the BMI was 26.4 kg/m² and 54.5% of the patients had a history of hypertension.¹⁷ In another Colombian study, a median age of 62 years, BMI of 26.7, SOFA score of 5, and mortality between 20% and 88% were found. Characteristics associated with mortality were age, use of vasopressors, and the need for renal replacement therapy.¹⁸

The most frequent clinical manifestations were cough and respiratory distress, which is consistent with the published literature.^9,13,19,20 The oxygenation disorder was severe, and the static compliance was low, this has been published in other studies and that in fact represents a higher risk of mortality and difficulty in ventilatory management due to the decrease in lung compliance and the requirements of deep sedation and neuromuscular relaxation.^19,21,22

Critically ill patients with COVID-19 require admission to the intensive care unit, for airway management, support with invasive mechanical ventilation and in more severe cases for assistance with extracorporeal oxygenation (ECMO).^13,23,24 In our work, invasive mechanical ventilation was used in most cases, with an average time of 7.1 days, and the most used mode was volume-assisted-controlled. Most of the patients were pronated, averaging 12 h a day for two days.

In a Mexican observational study, the use of invasive mechanical ventilation (IMV) was reported in 100% of the patients admitted to the ICU, this support was given for 11 days (median), about half of the patients were placed in prone position.¹⁵ In the SATICOVID study, the median IMV was 13 days.¹⁶ In both Latin American studies, the duration of invasive ventilation was longer than that found by us. In the multicenter study PRo-VENT-COVID, the volume-controlled mode was much lower than that found by us (19% vs. 90%), the most used ventilatory mode was pressure-control ventilation (52%) and Synchronized Intermittent Mandatory Ventilation (SIMV) in 7% of cases. Regarding the prone position, they reported its use in 25% of all patients and in 60% of patients with Pa/FiO₂ < 150, for a time of 8 h on average, which is below what was found in our study.²⁵

The organic dysfunction evaluated from prognostic scores such as SOFA or APACHE II has been associated with the severity of patients with COVID-19; in our study, we found an average of 7 points, which is higher than that found in other studies of our region^15,16 and in North American,²⁶ European²⁷ and Asian studies.²⁸

Mortality in our study was 60%, this is slightly higher compared to that reported in patients hospitalized in ICU in Argentina (57%) and Mexico (52%),^15,16 nevertheless, was higher than that reported in Canada (26%),²⁹ 30% in the United States of America,²⁶ Denmark (35–51%, depending on the weeks of ICU hospitalization),²⁷ 29% in London.¹⁹ This is possibly explained by the severity of the pulmonary involvement, most of the patients in our study showed a phenotype of high pulmonary elastance (phenotype H) and by the difference in the resources available to support very critical cases (e.g.: ECMO) between developed countries and some of the countries of our region.³⁰

Markers of severity and risk of mortality such as C-reactive protein, LDH, ferritin, advanced age, obesity, and kidney failure were like other studies.^{13,15,16,19,31} In the ventilatory aspect, patients with phenotype H, with higher PEEP, FiO₂ requirements and the need for pronation had a higher risk of mortality. From the above, we can interpret that the more severe the disease, the greater the risk of dying, therefore, these patients require a higher level of support.

Our work presents some important limitations, it is an observational study, of a single center, that although it was the largest hospitalization center for critical cases of COVID-19 in our region during the first year of the pandemic, this limits its generalizability.

Conclusions

COVID-19 is a disease with high morbidity and mortality, in our study, we found that a higher level of inflammation, impaired lung compliance, severe oxygenation disorder and the need for greater ventilatory support related to a higher risk of fatal outcome.

Conflict of interest

The authors declare they have no conflict of interest.