• Users Online: 497
  • Home
  • Print this page
  • Email this page
Home About us Editorial board Ahead of print Current issue Search Archives Submit article Instructions Contacts Login 


 
 Table of Contents  
ORIGINAL ARTICLE
Year : 2021  |  Volume : 13  |  Issue : 3  |  Page : 127-135

Determinants of intensive care unit admission of hospitalized patients with COVID-19 in Saudi Arabia: An analytic retrospective cohort


1 Department of Family and Community Medicine, University of Jeddah, Jeddah, Saudi Arabia
2 University of Tabuk, Tabuk, Saudi Arabia
3 King Saud Bin Abdulaziz University For Health Sciences, Riyadh, Saudi Arabia

Date of Submission27-Jun-2021
Date of Decision09-Apr-2021
Date of Acceptance30-Jul-2021
Date of Web Publication27-Sep-2021

Correspondence Address:
Dr. Sulafa Alqutub
Hamza Ibn AlQasim, AlSharafeyah, Jeddah 23218
Saudi Arabia
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/ijmbs.ijmbs_48_21

Rights and Permissions
  Abstract 


Background: In March 2020, Saudi Arabia (KSA) experienced a coronavirus disease 2019 (COVID-19) outbreak. The mitigation strategy aimed to reduce both the impact on vulnerable groups and the risk of admission to the intensive care unit (ICU). Risk factors, such as sequential organ failure, comorbidities, ventilation, and mortality, have not been described in different settings of care. Materials and Methods: A multicenter, retrospective chart review of 220 adults with COVID-19 admitted to the ICU included demographics and ICU admission factors (e.g., quick sequential organ failure assessment (qSOFA) score, ventilator status, comorbidities, days from laboratory confirmation to ICU admission, and days from hospitalization to ICU admission). Regression was utilized to identify predictors of need for mechanical ventilation (MV) and mortality in ICU patients. Results: ICU admission, COVID-19 hospital mortality, and ventilator-associated mortality rates were 26.5%, 44%, and 30.5%, respectively. The mean patients' age was 30 years. Across four cities, Jeddah patients were at the highest risk of MV (<0.001). Within the 1st day of hospitalization, without lymphocytopenia, non-Saudi patients with a qSOFA score of 2 and 3 were at the highest risk of ventilation (odds ratio [OR], 3.9; 95% confidence interval [CI], 1.72–8.66; OR, 11.4; 95% CI, 2.35–55.47; and OR, 6.1; 95% CI, 1.0–37.33, respectively). Moreover, within the same period of hospital stay, mechanically-ventilated patients with a qSOFA score of 3 who received antiviral medications were significantly at the higher risk of death (OR, 2.8.4; 95% CI, 1.44–5.64; OR, 13.1; 95% CI, 1.23–39.68; and OR, 2.2; 95% CI, 1.14–4.14, respectively). Conclusions: The 1st day of hospitalization, along with an assessment of the dyspnea status using the qSOFA score, is the window of opportunity for minimizing ICU admission risk. Neither lymphocytopenia nor comorbidities are associated with the risk of mechanical ventilation. Factors were also discussed. Reviews are needed on the indications for the use of antiviral agents, intubation, and ventilation in hospitalized patients.

Keywords: COVID-19, death, intensive care unit admission, Saudi Arabia, quick sequential organ failure assessment, ventilation


How to cite this article:
Alqutub S, Albalawi W, Alrajhi NN. Determinants of intensive care unit admission of hospitalized patients with COVID-19 in Saudi Arabia: An analytic retrospective cohort. Ibnosina J Med Biomed Sci 2021;13:127-35

How to cite this URL:
Alqutub S, Albalawi W, Alrajhi NN. Determinants of intensive care unit admission of hospitalized patients with COVID-19 in Saudi Arabia: An analytic retrospective cohort. Ibnosina J Med Biomed Sci [serial online] 2021 [cited 2021 Dec 8];13:127-35. Available from: http://www.ijmbs.org/text.asp?2021/13/3/127/326776




  Introduction Top


In the battle against the coronavirus disease 2019 (COVID-19) pandemic, Saudi Arabia (KSA) has adopted a mitigation strategy aimed at reducing the health impact on vulnerable groups and the risk of admission to the intensive care unit (ICU).[1] Risk factors for ICU admission, namely multiorgan failure, ventilation status, and associated mortality, have not been well described in patients with COVID-19 in the KSA.

Two studies have described the clinical characteristics and outcomes of hospitalized patients with COVID-19 in KSA. Both studies have revealed the association of COVID-19 with comorbidities at one center.[2],[3] However, the present study focused on national multicenter ICU admission rates and multiorgan failure based on quick sequential organ failure assessment (qSOFA) score, ventilator status, and death rate as secondary outcomes after ICU admission. The predictive value of the qSOFA score was significantly superior to that of SOFA for in-hospital mortality.[4]

The efficacy of hydroxychloroquine, alone or in combination with azithromycin, in reducing mortality among hospitalized patients with COVID-19 confirmed through polymerase chain reaction (PCR).[5],[6]

Halacli et al. demonstrated that the ICU mortality rate due to COVID-19 ranged from 16% to 78%.[7] Also, a single-center study of critically ill patients with COVID-19 concluded that endotracheal intubation and ventilation were associated with a mortality rate of 26%.[8] In addition, an ICU admission rate of 14.2% during hospital admission and an in-hospital mortality rate of 24.5% after receiving ventilation.[9]

We, therefore, aimed to provide insight into the differences in survival status between ventilated and nonventilated patients admitted to the ICU. Moreover, we aimed to assess the importance of the qSOFA score as a tool for predicting the severity of hospitalized patients with COVID-19, and hence, the risk of mortality in the ICU.


  Materials and Methods Top


Design and settings

This retrospective cohort chart review study of hospitalized patients with COVID-19 (confirmed through PCR) who required admission to the ICU during the pandemic from March to June 2020. Simple random sample recruitment from the assigned COVID-19 hospitals was performed until the desired sample size was reached. The pediatric patients and those with unknown or unclear outcomes were excluded. The Herra Hospital (Makkah) (with a variable ICU bed capacity of up to 15 beds), as well as Uhud hospital (Madinah), King Abdullah Medical complex (Jeddah), and Prince Mohammed bin Abdul-Aziz Hospital (Riyadh) with 70, 26, and 100 ICU beds, respectively, were included.

Sample size calculation

The sample size was calculated to ensure efficiency considering limited funding and strict criteria for accessing healthcare facilities during the pandemic. The assumption was based on the previously described clinical course of COVID-19; 14% of all diagnosed patients are at risk of developing severe disease symptoms and require admission to a hospital. Moreover, 5% of patients will become critically ill and require admission to the ICU.[4],[10],[11] The identified outcome variable in this study was admission to the ICU with or without ventilation. Open Epi info was used to determine the required sample size for this cohort study. The expected percentage of unexposed and exposed individuals with the intended outcome was 5.1% and 14%, respectively, at a 95% confidence interval (CI) with a power of 80%. The sample size determined using Kelsey methods was 172 cases/files. Considering death as an expected outcome for several patients admitted to the ICU and the possibility of limited access to some of the files, we increased the sample size by approximately 25% (i.e., 220 files) to compensate for the missing data or information.

Statistical analysis

Frequency distributions to describe categorical variables and the mean and standard deviation were used to report central tendency measures. For nonnormally-distributed continuous variables, median and interquartile ranges were used. Differences between groups were analyzed using the Chi-square test for categorical variables and the independent t-test for continuous variables. P < 0.05 denoted significant statistical differences. Binary logistic regression was applied to adjust for confounders. The Statistical Package for the Social Sciences software version 21 (IBM Corporation, Armonk, NY, USA) was used.

Data collection and variable definitions

Three trained data collectors extracted data from electronic health records. In Uhud Hospital, paper files were the primary source of data. Study variables collected on each patient included the following: (1) patient demographic variables: age in years, sex, nationality, occupation, city of hospitalization, duration of the hospital to ICU admission in days, and duration of confirmed positive PCR diagnosis to ICU admission; (2) clinical characteristics, including comorbidities (chronic obstructive pulmonary disease, diabetes mellitus (DM), cardiovascular disease, hypertension, and other chronic illnesses including chronic kidney diseases), presence of lymphopenia, and use of antimalarial, antibacterial, and antiviral agents. In addition, data on ICU ventilator status or use, qSOFA score at the time of admission to the ICU, and mortality as an event were also collected.

SOFA assessment includes a Glasgow coma scale score of ≤13, systolic blood pressure of ≤100 mm Hg, and respiratory rate of ≥22/min (1 point each; score range, 0–3 points).[10] Thus, a qSOFA score of 0 denotes no organ failure. In contrast, qSOFA scores of 1, 2, and 3 denote the failure of 1, 2, and 3 organs, respectively. The qSOFA score is a practical and accessible approach for sepsis screening.[4] In the present study, the qSOFA score was documented by trained ICU nurses and doctors within 24 h of admission to the ICU.

Lymphocytopenia was defined as a total lymphocyte count of <1000/mL (<1 × 109/L) in adults, where the normal lymphocyte count is 1000–4800/mcL.[12] To assure the reliability of data reporting across the four research sites, dichotomization was applied based on the reported low laboratory readings. Lymphocytopenia was reported twice: at hospital admission and ICU admission.

Antimalarial agents included chloroquine and/or hydroxychloroquine administered to the sample population. Antiviral agents included all antiviral agents administered to the sample population, including oseltamivir. Antibiotics included all antibacterial agents administered to the sample population, including azithromycin.

A pilot study was conducted in the selected hospitals. The study included 17 charts, and adjustment was carried out to assure the feasibility and accuracy of the data collection sheet.


  Results Top


Intensive care unit admission rate

From March 1, 2020, to May 31, 2020, out of 771 hospitalized patients with PCR-confirmed COVID-19 in Makkah and Jeddah, 204 patients required admission to the ICU (26.5%).

Demographic characteristics and in-hospital events

The present study included a sample of 220 patients from four different cities. The mean (standard deviation) age of the participants was 30 (13.1) years. The majority were males and non-Saudi [Table 1]. Sequences of in-hospital events are shown in [Figure 1].
Figure 1: Sequence of in-hospital clinical events

Click here to view
Table 1: Demographic characteristics of patients with COVID-19 admitted to the intensive care unit in four cities (n=220)

Click here to view


Lymphocytopenia on hospital admission was found in 42% of patients, while on ICU admission was found to be at 66%. Antiviral medication was given to 36% of patients, while antimalarial and antibacterial medications were given to 36% and 63%, respectively. Of the valid 171 cases, only 72 reported lymphocytopenia at the time of hospital admission (33%). Of the 219 patients, 144 reported lymphocytopenia at admission to the ICU (66%).

One hundred and twenty-one patients (56%) required ventilation, and 96 patients (44%) died. Slightly higher death rate was observed among patients using ventilators (n = 67; 30.5%). While the survival rate among patients who did not use ventilators was (n = 68; 31%). The qSOFA scores are shown in [Table 2].
Table 2: Clinical characteristics of patients with COVID-19 admitted to the intensive care unit in four cities (n=220)

Click here to view


Factors associated with ventilation status in univariate analysis

The odds of being ventilated were significantly higher among non-Saudis (odds ratio [OR] =2.6, 95% confidence interval [CI] 1.41–4.98), those who had no lymphocytopenia on hospital admission (OR = 2.0, 95% CI 1.08–3.85), and those who had no lymphocytopenia on ICU admission (OR = 2.1, 95% CI 1.18–3.69). The odds of being ventilated were also significantly higher among those who did not receive antiviral medications (OR = 2.0, 95% CI 1.15–3.57), those who had no antimalarial administration (OR = 2.3, 95% CI 1.30–4.06), and those who had received antibacterial medication (OR = 2.9, 95% CI 1.6–5.24). Those who had a qSOFA score of 2or 3 were more likely to be ventilated than those who had a 0 (OR = 11.3, 95% CI 2.54–49.9 and OR = 7.5, 95% CI 1.38–40.24, respectively). Median hospital stay from admission to ICU admission was significantly shorter among those who had been ventilated (P = 0.001). On the other hand, the median duration from confirmed PCR diagnosis to ICU admission was longer among those who had been ventilated (P = 0.010). Across the cities, the highest ventilation rate was found among those from Jeddah (100%) [Table 3].
Table 3: Factors associated with ventilation status in univariate analysis

Click here to view


Prediction of ventilation status in multivariate analysis

All factors associated with ventilation status in univariate analysis were entered into the multiple logistic regression analysis. Significant predictors of ventilation status in the model were: being non-Saudi (P = 0.001), having no lymphocytopenia on ICU admission (P = 0.020), and having a qSOFA score of 2 or 3 (P = 0.002, P = 0.049, respectively). The total model was significant and explained 35% of the outcome variance [Table 4].
Table 4: Predictors of ventilation status in multiple logistic regression

Click here to view


Factors associated with mortality in univariate analysis

The rate of death was significantly higher among ventilated cases (OR = 3.1, 95% CI 1.76–5.53), those who were given antiviral medication (OR = 1.9, 95% CI 1.08–3.3), and those who had a qSOFA score of 2 or 3 (OR = 10.6, 95% CI 1.2–88.3; OR = 28.8, 95% CI 2.8–56.4, respectively). Median stay from hospital admission to ICU admission was short among deceased cases (P = 0.002). The highest survival rate was in Riyadh (100%). and, the highest death rate was in Jeddah (77.1%) [Table 5].
Table 5: Factors associated with mortality in univariate analysis

Click here to view


Prediction od mortality in multivariate analysis

All factors associated with mortality in univariate analysis were entered into the multiple logistic regression. Significant predictors of mortality in the model were: antiviral administration (P = 0.018), being ventilated (P = 0.002), and having a qSOFA score of 3 (P = 0.033). The total model was significant and explained 20% of the variance of the outcome [Table 6].
Table 6: Predictors of mortality in multiple logistic regression

Click here to view



  Discussion Top


Within 1 day of median hospital stay, non-Saudi patients with qSOFA scores of 2 and 3, without lymphocytopenia, were at the highest risk of admission to the ICU and hence ventilation. Moreover, within the same period of hospital stay, mechanical ventilation (MV) patients with a qSOFA score of 3 who received antiviral medications were significantly at higher risk of death from COVID-19. Our study showed that MV was associated with mortality among patients with COVID-19 admitted to the ICU at 30.5%. The association of mortality with ventilation in this study was higher than that reported in a hospital epidemiology study performed in the United States of America (USA).[9]

Regarding overall in-hospital mortality, our results revealed an ICU mortality rate of 44% among patients with COVID-19. In a single-center study of critically ill patients with COVID-19 in Italy, endotracheal intubation and ventilation were associated with a mortality rate of 26%.[8] Another case series from 12 hospitals in New York showed a death rate of 24% in hospitalized patients.[9] The higher death rate reported in the present study may be attributable to ventilators and variability in practice among the multiple centers involved. Moreover, clinical factors, such as coagulopathies, are proposed to be associated with an increased rate of in-hospital mortality among ventilated patients.[15] However, the potential roles of these factors were beyond the scope of the present study.

During the peak of the pandemic in Guangzhou, China, there were a total of 19,425 patients who required hospitalization, 2087 patients required critical care, and the approximate rate of ICU admission was 11%.[13],[14] The combined rate of ICU admission in Jeddah and Makkah was 26.5% during the study period; this rate was two-fold higher than that reported in Guangzhou. The differences in ICU admission rates between the KSA and China may be attributable to the differences in the pandemic's peak during the study period and the associated comorbidities.[15]

Through a systematic review and meta-analysis, Jiang et al. examined the role of qSOFA in predicting mortality. The results revealed a strong association of mortality with a qSOFA score >2 and the poor sensitivity of this score for the early prediction of mortality in patients with pneumonia.[15] The present study on patients with severe COVID-19 demonstrated that those with a qSOFA score of 1 had the highest survival rate (n = 76; 62%) [Table 5]. Moreover, a qSOFA score of 3 was a predictor of mortality among patients with COVID-19 [Table 6]. Our findings were aligned with a previous study on the relationship between mortality and the qSOFA score.[15]

Patients with a qSOFA score >1 were significantly more likely to be ventilated, particularly in Jeddah. This finding is attributed to multiorgan failure as an indicator for intubation in the Saudi Ministry of Health Protocol on MV for COVID-19.[16] Jain et al. conducted a meta-analysis of 1813 patients with COVID-19. They reported that disease severity was determined by dyspnea, followed by cardiovascular diseases and hypertension. Older age males were strong demographic predictors of disease severity.[17]

In a systematic review, Jain et al. concluded that patients with dyspnea had a 6.6-fold higher risk of ICU admission than those without dyspnea.[17] Training of staff on the proper use of the qSOFA score, including dyspnea as an essential component, inpatient assessment. Tachypnea indicates early respiratory failure; hence, assessment within the 1st day of admission to the hospital will facilitate earlier identification of risk, thereby allowing for secondary prevention.

The present study included a young population, non-Saudi, and most were male patients. Furthermore, comorbidities were neither predictors for ventilator use nor mortality compared with patients without chronic diseases. Our findings were inconsistent with those of a recently published study conducted by Lei et al., which showed higher ICU admission odds among patients with DM versus nondiabetics. Moreover, older patients with DM respiratory rate >24 and hemoglobin A1C levels >7% were at risk of admission to the ICU. Details on the reasons for this risk were discussed by Lei et al.[18] Despite the relatively high prevalence of DM in the KSA,[5] our results are incomparable because of the differences in the demographic nature of the study group.

Our findings showed that patients who received antiviral agents were at a higher risk of death than those who did not. This is consistent with notions about moderate to very low certainty of evidence on the effectiveness of various antiviral agents such as remdesivir in reducing death among severe and critically ill COVID-19 patients.[19],[20]

There are a few limitations. First, the inherent nature of record review and the inability to collect relevant data on some determinants (body mass index, vaccination status, D-dimer, and fibrinogen levels). Second, Saudi nationals were underrepresented. Third, the relatively small sample size and self-funding, the inclusion of patients from four different cities, are also possible limitations. Finally, some of the interventions applied to suppress a cytokine storm (e.g., corticosteroids) were not tested in the study.


  Conclusions Top


In severe COVID-19 cases, the rates of ICU admission, ICU mortality, and ventilator-associated mortality in the studied settings are 26.5%, 44%, and 30.5%, respectively. The 1st day of hospital admission is a critical period for determining the decision to admit patients to the ICU. MV is a strong determinant of ICU admission and consequently qSOFA scores of 2 and 3. Ventilation and antiviral administration are associated with death.

Longitudinal studies need to investigate the determinants of ICU admission and survival status including private hospitals in the KSA. To provide information regarding the variability in the indications for admission to the ICU and differences in outcomes.

Authors' contribution

All authors contributed substantially to the conception, data collection and analysis, and drafting of the manuscript. They all approved the final version of the manuscript.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

Compliance with ethical principles

The study was approved by the institutional review board of King Fahad Medical City. (KACST: KSA H-01-R-012). The study was exempted from requiring consent.



 
  References Top

1.
Ferguson NM, Laydon D, Nedjati-Gilani G, Imai N, Ainslie K, Baguelin M, et al. Impact non-pharmaceutical interventions (NPIs) to reduce COVID- 19 mortality and healthcare demand. 2020.  Back to cited text no. 1
    
2.
Almalki ZS, Khan MF, Almazrou S, Alanazi AS, Iqbal MS, Alqahtani A, et al. Clinical characteristics and outcomes among COVID-19 hospitalized patients with chronic conditions: A retrospective single-center study. J Multidiscip Healthc 2020;13:1089-97.  Back to cited text no. 2
    
3.
Alsofayan YM, Althunayyan SM, Khan AA, Hakawi AM, Assiri AM. Clinical characteristics of COVID-19 in Saudi Arabia: A national retrospective study. J Infect Public Health 2020;13:920-5.  Back to cited text no. 3
    
4.
Schrag A, Rubenfeld G, Kahn JM, Shankar-Hari M, Singer M. Assessment of clinical criteria for sepsis for the third international consensus definitions for sepsis and septic shock (Sepsis-3). JAMA 2016;315:762-74.  Back to cited text no. 4
    
5.
Arshad S, Kilgore P, Chaudhry ZS, Jacobsen G, Dee D, Huitsing K, et al. Treatment with hydroxychloroquine, azithromycin, and combination in patients hospitalized with COVID-19. Int J Infect Dis 2020;97:396-403.  Back to cited text no. 5
    
6.
Saudi MOH. Coronavirus Disease 19. Vol. 1. Saudi MOH; 2020. p. 19.  Back to cited text no. 6
    
7.
Article R. Critically ill COVID-19 patient. 2020;585-91. [doi: 10.3906/sag-2004-122].  Back to cited text no. 7
    
8.
Grasselli G, Zangrillo A, Zanella A, Antonelli M, Cabrini L, Castelli A, et al. Baseline characteristics and outcomes of 1591 patients infected with SARS-CoV-2 admitted to ICUs of the Lombardy Region, Italy 2020. JAMA 2020. [doi: 10.1001/jama. 2020.5394].  Back to cited text no. 8
    
9.
Richardson S, Hirsch JS, Narasimhan M, Crawford JM, Mcginn T, Davidson KW. Presenting characteristics, comorbidities, and outcomes among 5700 patients hospitalized with COVID-19 in the New York City Area. JAMA 2020;323:2052-9.  Back to cited text no. 9
    
10.
Feist B. Screening for sepsis: SIRS or qSOFA? A literature review. Emerg Nurse 2019;27:13-8.  Back to cited text no. 10
    
11.
Available from: Https://WwwUptodateCom/Contents/Coronavirus-Disease-2019-Covid-19n.d. [Last accessed on 2021 Sep 04].  Back to cited text no. 11
    
12.
MSD n.d. Available from: https://www.msdmanuals.com/professional/hematology-and-oncology/leukopenias/lymphocytopenia. [Last accessed on 2021 Sep 04].  Back to cited text no. 12
    
13.
14.
Li R, Rivers C, Tan Q, Murray MB, Toner E, Lipsitch M. Estimated demand for US hospital inpatient and intensive care unit beds for patients with COVID-19 based on comparisons with wuhan and Guangzhou, China. JAMA Netw Open 2020;3:e208297.  Back to cited text no. 14
    
15.
Jiang J, Yang J, Jin Y, Cao J, Lu Y. Predictive symptoms and comorbidities for severe COVID-19 and intensive care unit admission: a systematic review and meta- analysis. International Journal of Public Health 2020;65:533-46.  Back to cited text no. 15
    
16.
Background I. Mechanical Ventilation Protocol for COVID-19 [Internet]. Vol. 2019. 2020. Available from: https://www.moh.gov.sa/en/Ministry/MediaCenter/Publications/Pages/covid19.aspx. [Last accessed on 2021 Sep 04].  Back to cited text no. 16
    
17.
Jain V, Yuan J. Predictive symptoms and comorbidities for severe COVID-19 and intensive care unit admission: A systematic review and meta-analysis. Int J Public Health 2020;65:533-46.  Back to cited text no. 17
    
18.
He F, Quan Y, Lei M, Liu R, Qin S, Zeng J, et al. Clinical features and risk factors for ICU admission in COVID-19 patients with cardiovascular diseases. Aging Dis 2020;11:763-9.  Back to cited text no. 18
    
19.
Liu W, Zhou P, Chen K, Ye Z, Liu F, Li X, et al. Efficacy and safety of antiviral treatment for COVID-19 from evidence in studies of SARS-CoV-2 and other acute viral infections: A systematic review and meta-analysis. CMAJ 2020;192:E734-44.  Back to cited text no. 19
    
20.
Al-abdouh A, Bizanti A, Barbarawi M, Jabri A, Kumar A. Since January 2020 Elsevier has created a COVID-19 resource centre with free information in English and Mandarin on the novel coronavirus COVID- 19 . The COVID-19 resource centre is hosted on Elsevier Connect, the company's public news and information. 2020.  Back to cited text no. 20
    


    Figures

  [Figure 1]
 
 
    Tables

  [Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]



 

Top
 
 
  Search
 
Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

 
  In this article
Abstract
Introduction
Materials and Me...
Results
Discussion
Conclusions
References
Article Figures
Article Tables

 Article Access Statistics
    Viewed640    
    Printed22    
    Emailed0    
    PDF Downloaded55    
    Comments [Add]    

Recommend this journal


[TAG2]
[TAG3]
[TAG4]