High prevalence of acute stress disorder and persisting symptoms in ICU survivors after COVID-19

Dear Editor,

Post-traumatic stress disorder may surge following exposure to life-threatening events and affects around 20% of intensive care unit (ICU) survivors [1, 2]; a higher rate (38.8%) was observed in Middle East and Severe Acute Respiratory Syndromes’ outbreaks [3]. The coronavirus disease 19 (COVID-19) pandemic [4] presents all the features to deeply impact not only on physical but also on mental health: patients are isolated from their families, communication with healthcare providers is difficult because of personal protective equipment and discrepancy between patients’ number and staff members is evident. A significant psychological impact on general population and healthcare providers was demonstrated [5].

In a follow-up hospital visit, 47 patients (males 70.2%, 59 [50–66] year-old) admitted to our ICU for COVID-19 from February 21st to June 5th, 2020 (Fig. 1 ESM) were assessed in person by an intensivist at least 1 month after hospital discharge [median time 72 [55–92] days). Informed consent was collected following the ad hoc procedures defined by the Ethics Committee for COVID-19 pandemic. The Impact of Event Scale–Revised (IES-R–Table 1 ESM) was used to screen for acute stress disorder (ASD), defined by a score ≥ 33. Median IES-R score was 30 [19–41]; 19 patients (40.4%) showed ASD. Among subscales, intrusion average score was higher than avoidance (1.6 [1–2.4] vs. 1.25 [0.8–1.8]; p = 0.0660); both were higher than hyperarousal (0.9 [0.3–1.3]; p < 0.0001 and p = 0.0114). No significant difference in patients’ characteristics, length of stay in ICU/hospital, need and length of invasive/non-invasive ventilation and drugs was observed comparing ASD and non-ASD patients (Table 1). ASD patients were less hypoxemic at ICU admission; we can just speculate that, being less dyspnoeic, they possibly had more clear-headed experiences of their own condition and of the surrounding environment, including other patients’ struggle and death. Patients with IES-R score higher than 30 were addressed to outpatients’ facilities of their catchment area, to receive an accurate diagnosis and psychotherapy/medication, if necessary.

Table 1.

Features of the patients discharged from ICU after COVID-19 with and without acute stress disorder

	Overall (n = 47)	No ASD (n = 28)	ASD (n = 19)	p value
Age–years, median [IQR]	59 [50–66]	60.5 [50.5–67]	57 [50–65]	0.4540
Sex–males, n (%)	33 (70.2)	22 (78.6)	11 (57.9)	0.195
BMI–kg/m2, median [IQR]	27.7 [25.4–30.7]	27.3 [25.4–31]	27.7 [25.1–30.7]	0.5543
SAPS II–median [IQR]	31 [27–45]	34 [26–48]	30.5 [27–39.5]	0.3109
Past medical history of psychiatric diseases–n (%)	2 (4.3)	0 (0.0)	2 (10.5)	0.079
FiO2 at ICU admission—median [IQR]	0.8 [0.6–1]	1 [0.7–1]	0.6 [0.6–1]	0.0734
PaO2/FiO2 at ICU admission–mmHg, median [IQR]	123 [84–159]	100 [82–130]	142 [121–195]	0.0114
Length of stay in hospital before ICU–days, median [IQR]	3 [1–8]	3 [1.5–7.5]	3 [0–8]	0.6942
Length of stay in ICU–days, median [IQR]	24 [10–37]	25.5 [11–40.5]	19 [9–37]	0.2828
Length of stay in hospital after ICU–days, median [IQR]	9 [6–15]	9 [6–14]	12 [6–21]	0.4712
Length of CPAP before ICU admission–days, median [IQR]	2 [1–5]	3 [1–5]	1 [1–8]	0.6927
Invasive mechanical ventilation in ICU–n (%)	34 (72.3)	21 (75)	13 (68.4)	0.621
Length of IMV–days, median [IQR]	17.5 [9–28]	17 [11–30]	18 [7–25]	0.3293
Length of NMBA–days, median [IQR]	7 [3.5–11.5]	7 [6–12]	5 [3–11]	0.2163
Pronation–n (%)	19 (40.4)	11 (39.3))	8 (42.1)	0.773
Benzodiazepine sedation in ICU–n (%)	23 (48.9)	16 (57.1)	7 (36.8)	0.238
Dexmedetomidine sedation in ICU–n, (%)	23 (48.9)	13 (46.4)	10 (52.6)	0.770

ASD acute stress disorder; IQR Interquartile range; BMI Body mass index; SAPS II Simplified acute physiology score; CPAP continuous positive airways pressure; ICU intensive care unit; PEEP positive end-expiratory pressure; PaO2 arterial partial oxygen pressure; FiO₂ fraction of inspired oxygen; IMV invasive mechanical ventilation; NMBA neuromuscular blocking agents

Anosmia, ageusia, altered hearing and balance persisted in 10.6%, 19.2%, 10.6% and 19.2%, respectively. Asthenia and dyspnoea were scored with a 1–5 range. Asthenia persisted in 97.8% of patients (median score 2 [1–3]). Four patients (8.5%) complained of dyspnoea at rest, two required home oxygen therapy. Patients complaining of dyspnoea for mild, moderate and severe efforts were 13 (27.7%), 46 (97.8%) and 46 (97.8%), respectively. The number of patients complaining of dyspnoea in each of these classes did not differ significantly in ASD and non-ASD patients; however, dyspnoea score was higher in ASD patients (Table 2 ESM).

Most of the patients had unclear memories of ICU stay; the most common complaint was the feeling of complete isolation due to limited possibility to communicate with families/physicians. Many patients described nightmares, feeling of derealization and the initial belief of being in a fake hospital; this mainly affected the first patients of the pandemic. 10 patients (21.3%–ASD 4 (21.1%) vs. non-ASD patients 6 (21.4%); p = 0.975) wanted to see the ICU, both to meet the ICU staff and to help their memories’ reconstruction process.

In conclusion, ASD prevalence is high among ICU survivors after COVID-19; a systematic screening should be performed to reduce chronic effects of critical illness. Symptoms of asthenia and dyspnoea for moderate efforts persist for weeks after ICU and hospital discharge in a large majority of the patients.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

SM, GS, GT, FM actively contributed to the conception of the study, data collection and analysis; all the authors actively contributed to the redaction of the draft, provided critical revision of the article and provided final approval of the version submitted for publication.

Funding

Institutional funding.

Compliance with ethical standards

Conflicts of interest

FM received fees for lectures from GE Healthcare, Hamilton Medical, SEDA SpA, outside the present work. SM received fees for lectures from GE Healthcare, outside the present work. GT received fees for lectures from GE Healthcare, outside the present work. A research agreement is active between University of Pavia and Hamilton Medical. The other authors declare no conflict of interest.

Informed consent

All patients signed an informed consent for data publication.

Ethics approval

Approved by local ethic committee.

Consent to participate

All patients signed an informed consent for data collection.