Comorbidities and Risk Factors for Severe Outcomes in COVID-19 Patients in Saudi Arabia: A Retrospective Cohort Study

Introduction

The Ministry of Health (MOH) of the Kingdom of Saudi Arabia (KSA) reported 4 new cases of Covid-19 on March 9, 2020, 7 days after the first positive case was identified.¹ COVID-19 infection occurs from airborne exposure to the virus, particularly through droplets/aerosols during coughing or sneezing and physical contact with infected person and objects^5,6 via nasal, oral or fecal route² spreading quickly via infected travelers. The first case in KSA was reported in a traveler from Iran to Qatif^3,4 and confirmed COVID-19 incidences rose to 3717 as of June 10, 2020 with over 308,600 cases and 3600 deaths estimated by August 24, 2020.^5,6 So far, KSA has the highest COVID-19 mortality rate in the Gulf Cooperation Council (GCC) region after Iran and Iraq, and has the third highest positive cases in the Middle East region.⁷ Riyadh has the highest number of residents and citizens infected with SARS-CoV-2 in KSA (https://covid19.moh.gov.sa/).

Several complications are associated with COVID-19 infections, which enhance the morbidity and mortality rates.⁸ Previous studies have identified poor/weak health-care systems and failure to adhere to precautionary measures as high contributory factors to infection rates especially in poor to low-income countries.^9,10 High risks of severity and mortality have been widely reported in older persons infected with COVID-19, a finding that has been consistent in several countries.^11–13 Surgical procedures especially transplants and pre-existing comorbidities, such as hypertension, obesity, diabetes mellitus (DM), cardiovascular disease, respiratory diseases, malignancies, and other chronic, non-communicable diseases reportedly worsen clinical outcomes in COVID-19 patients, thereby increasing fatality rates.^14,15

Current global data revealed a 4.8% case fatality ratio (CFR) in China, 5.8% in the USA, 3.5% in Italy, 2.4% in France, 2.7% in the United Kingdom, 2.5% in Brazil, and 1.7% in KSA (https://ourworldindata.org/covid-mortality-risk). About 348 new cases are currently being recorded daily in KSA and total deaths currently stand at 6869 (retrieved 04-24-2021 from Data https://github.com/CSSEGISandData/COVID-19). Decreasing CFR in KSA could be due to multiple swift measures introduced by KSA’s government to mitigate the spread of the virus or the recent exposures to MERS-CoV epidemic.^16,17

In an early data (March 2020) involving 1519 confirmed COVID-19 cases with average mean age of 36 years from KSA, hypertension was the most common comorbidity with a 8.8%, followed by Diabetes Mellitus (DM) with a 7.6% rate.¹⁸ Notwithstanding, despite the increases in fatality rates among COVID-19 patients in KSA, clinical data on contributory risk factors are insufficient. Therefore, this present study aims to describe the demography, clinical characteristics and comorbidities of infected patients, and identify the risks they contribute to the severity and mortality rates for the disease.

Methodology

Study Population

In this retrospective study, all patients above 18 years old who were COVID-19 positive and hospitalized at Prince Mohammed Bin Abdulaziz Hospital (PMAH), Riyadh, KSA from May to August 2020 were included. PMAH is a government-owned 500-bed hospital situated in the eastern part of Riyadh and one of the largest referral hospitals in KSA providing secondary healthcare to various acute or chronic medical conditions that excludes pediatric and gynecological. It houses 2 emergency, 36 medical and 9 intensive care units. These were isolated via HEPA filters during the COVID-19 pandemic to provide emergency services for diagnosed COVID-19 patients. A positive Covid-19 result was determined by reverse-transcriptase-polymerase-chain-reaction (RT-PCR) assay of specimens collected by a nasopharyngeal swab. The diagnoses followed guidelines approved by the Saudi Center for Disease Prevention and Control. This study has been approved by the Institutional Review Board (IRB) managed by the Deanship of Scientific Research at Imam Abdulrahman Bin Faisal University in Dammam, KSA.

Data Collection and Analysis

The retrieved clinical information included patient’s demographics, BMI, and comorbidities, which encompass hypertension, DM, cardiovascular, respiratory, renal, neurological, liver diseases, cancer and other diseases. Data on hospital’s Intensive Care Unit (ICU) admission and non-admission were also collected to determine the disease severity. Ages were categorized into seven groups; 20–29; 30–39; 40–49; 50–59; 60–69; 70–79 and >79 years while the BMIs were classified into three groups; ≥30, ≥35 and ≥40.

Analyses were then performed using Python Pandas. Categorical variables evaluated in this study were compared using the Chi Square test. Demographic data in addition to patients’ comorbidities were presented as percentages, median and interquartile range (IQR) while the Kaplan-Meier estimator was used to evaluate the survival functions of the different variables. Matplotlib was used for visualization, SciPy for performing Chi Square test, and lifelines for survival analysis. Univariate and multivariate logistic regression methods were used to adjust for the effects of patient’s age, gender, comorbidities and the outcomes. P-values ≤0.05 were considered significant.

Results

A total of 565 patients hospitalized with COVID-19 at Prince Mohammed Bin Abdulaziz Hospital, Riyadh, KSA were inducted in the study. Patients were stratified based on their ICU admission and mortality rates (Table 1). The majority of the patients were males (75.6%) while distribution based on nationality further revealed that 79.7% of the males were non-nationals (Table 1).

Table 1 Patients’ Characteristics

From the total 565 patients, 30.4% had a BMI greater than 30, 14.0% had a BMI greater than 35 while 6.9% had a BMI greater than 40. Comorbidities were present in most of the patients, with respiratory diseases being the most common (46.9%), followed by DM (41.6%). Other reported comorbidities were hypertension (33.8%), cardiovascular (17.0%), renal (14%), neurological (7.3%), cancer (1.8%), and liver (0.4%) diseases. Most patients in the older age-groups required ICU admission which indicated increased disease severity among this group. Requirement for ICU admission was more prevalent in males compared to females (18.2% vs 16.8%) while most patients with BMI ≥ 30 needed ICU admission compared to those with BMI ≥ 35 and BMI ≥ 40 (15.1% vs 13.9% vs 7.8%). A higher proportion of non-Saudi nationals were admitted to the ICU relative to Saudi nationals (18.7% vs 15.3%). However, none of the patients’ characteristics significantly determined the need for ICU admission. There was no significant difference in COVID-19 patients with hypertension, DM, cancer, neurological, liver and other comorbidities that required intensive care compared to patients which did not. Relatively, there was a significant need in COVID-19 patients with renal, cardiovascular and respiratory diseases for ICU admission (p<0.001) indicative of higher severities in patients with these comorbidities.

Kaplan-Meier survival analysis curve showed a trend that survival time was shortest in patients in age-group 50–59 years, although not significantly different from other age-groups (Figure 1A–E). Similarly, there were no significant differences in the survival time between Saudi nationals and non-nationals coupled with persons having BMIs <40 and ≥40. Although the trend for survival time seemed to be shorter in females, it is also not significantly different from males (Figure 2B–E). However, survival time is significantly shortened in patients admitted at the ICU (p<0.001). Patients with cardiovascular, respiratory and renal comorbidities had significantly shorter survival time (p<0.001) than those with hypertension, DM and other diseases (Figure 2A–I). Neurological comorbidities also showed increased mortality rates in COVID-19 patients (p=0.04). The overall mortality rate is 11.2% and as observed, none of the patients’ characteristics had significant effects on the rate of deaths among the COVID-19 patients. Likewise, there was no significant difference in mortality rates among patients with hypertension, DM, cancer, neurological and liver diseases. However, a higher number of patients with cardiovascular, respiratory and renal diseases died compared to those with other comorbidities (p<0.001). Also, we noted a significant difference in the mortality rate of patients with neurological comorbidities (p=0.04).

Figure 1 Kaplan Meier survival function analyses of the patients’ characteristics investigated for (A) Age groups, (B) Gender, (C) ICU/non-ICU, (D) BMI, and (E) Nationality.

Figure 2 Continue.

Figure 2 Kaplan Meier survival function analyses of the comorbidities; (A) Hypertension, (B) Diabetes, (C) Cardiovascular diseases, (D) Respiratory diseases, (E) Renal diseases, (F) Cancer (G) Neurological diseases, (H) Weight, and (I) Others.

Univariate cox proportional hazards regression model revealed that age [HR=1.3 (CI 1.1–1.5)], BMI ≥ 40 [HR=0.2 (CI 0.0–1.5)], nationality [HR=1.3 (CI 0.7–2.2)], and gender [HR=0.7 (CI 0.4–1.3)] were not significant risk factors associated with death among the hospitalized COVID-19 patients. Likewise, hypertension [HR=1.1 (CI 0.7–1.9)], diabetes [HR=1.1 (CI 0.7–1.9)], renal [HR=2.4 (CI 1.4–3.9)], cancer [HR=2.4 (CI 0.9–6.5)], neurological [HR=1.3 (CI 0.7–2.4)], and other [HR=1.3 (CI 0.7–2.5)] comorbidities do not present any significant risk to patients’ fatalities (Table 2, Figures 3A, B, D, E, H and I and 4A–D). Comparatively, mortality rates significantly increased among ICU admissions [HR=4.2 (CI 2.5–7.2); p<0.005], patients with cardiovascular diseases [HR=4.1 (CI 2.5–6.7); p<0.005] or respiratory diseases [HR=4.0 (CI 2.0–8.1); p<0.005] (Figure 3C, F and G). Multivariate cox proportional hazards regression model further emphasized cardiovascular disease [HR=2.9 (CI 1.7–5.0); p<0.005] as a major risk factor that significantly contributed to death rates among the hospitalized COVID-19 patients (Figure 5A). This was also supported by the Kaplan Meier Survival function analyses plot for the most significant risk factors (ICU admission, cardiovascular and respiratory diseases) relative to other variables (Figure 5B).

Table 2 Variables for the Outcomes of Interest Using Univariate and Multivariate Cox Proportional Hazards Regression Model

Figure 3 Univariate analyses with forest plots presenting the HR and 95% CI for mortality of COVID-19 based on the patient’s characteristics and comorbidities present. (A) Hypertension and COVID-19 mortality, (B) Age and COVID-19 mortality, (C) ICU and COVID-19 mortality (p=0.0010), (D) BMI>40 and COVID-19 mortality, (E) diabetes and COVID-19 mortality, (F) respiratory disease and COVID-19 mortality (p=0.004), (G) cardiovascular disease and COVID-19 severity (p<0.001) (H), renal disease and COVID-19 mortality, (I) cancer and COVID-19 mortality.

Figure 4 Univariate analyses with forest plots presenting the HR and 95% CI for mortality of COVID-19 based on the patient’s characteristics and comorbidities present. (A) Neurological diseases and COVID-19 mortality, (B) Other diseases and COVID-19 mortality, (C) Gender and COVID-19 mortality, (D) Nationality and COVID-19 mortality.

Figure 5 Forest plot from multivariate Cox hazards regression model presenting significant variables, (A) HR and 95% CI for COVID-19 mortality. (B) Kaplan Meier Survival function analyses for the top three significant variables compared to other factors.

Discussion

This study investigates the demographics, characteristics and comorbidities of COVID-19 patients hospitalized at Prince Mohammed Bin Abdulaziz Hospital, Riyadh, KSA as risk factors or predictors that may contribute to disease severity (defined by admission at the ICU) and death rate. We aimed at identifying the correlation between these variables and how they influence clinical outcomes as previously enumerated in KSA. Hence, to the best of our knowledge, this is first of few studies that succinctly describe the characteristics of a large cohort of hospitalized COVID-19 patients in Riyadh, KSA.

In this study, 75.75% were males compared to the remaining 24.25% that were females, in agreement with recent studies,^19–21 and in contrast to some other studies which showed that females were predominantly infected by SARS-CoV-2.^22,23 High susceptibility among males could be due to the high number of men (Saudi and non-Saudi) in the Saudi population including expatriates, which has a more prevalent male count than female. Recent studies have linked low susceptibility in females to variations in innate immunity, steroid hormones and other factors associated with sex chromosomes.²⁴ For instance, reduced testosterone levels have been widely linked to high severity in males²⁵ The regulatory roles played by androgen on transmembrane serine protease 2 (TMPRSS2) coupled with the positioning of the angiotensin converting enzyme 2 (ACE2) gene on X-chromosome underlie increased ACE2 levels or rebalancing of ACE1/ACE2, which altogether, could account for favorable outcomes in females infected with COVID-19.²⁶ In addition, older males have the obligations to provide for their families, hence they constituted majority of the working population in jobs, such as construction, logistics, delivery and other essential services during the COVID-19 outbreak which increased their contacts and interactions within the society. On the contrary, majority of jobs performed by females enabled them to remain at home and work remotely, including a proportion who are housewives. Moreover, the younger male population have higher tendencies of engaging in more social, sporting and cultural activities which may involve large gatherings. Also common to both young and adult males are social lifestyles or habits, such as smoking compared to females. These socio-cultural factors further account for high susceptibility in the male gender and could therefore shed more light on findings from this study as more males required ICU admission. However, death rate was higher in Saudi females than Saudi men, and other non-Saudi males and females. This is an interesting observation as it contradicts current global assertions on COVID-19 mortality rates in females as reported in earlier epidemiological and population-based studies.^27–31

In this present study, estimated median age for patients was 52 years which was higher than previous studies (44, 36 or 37 years) among infected COVID-19 patients in KSA,^16,18,32,33 while some others reported median ages between 37 and 70.5.^34,35 Our findings, however, contradict with some previous non-Saudi studies that suggested ages above 50 and 60 years as high-risk factors.^35–37 Earlier studies reported that relative to the rest of the world, COVID-19 affects younger persons in KSA.^16,18 Such incidences were observed in our study as COVID-19 cases were prevalent in persons with ages between 30 and 69 years. However, the infection seems to have occurred across our age-groups that ranged from 19 to 102 years indicating disease susceptibility across these ages. High infection and severity rates in adult persons between 50 and 59 years as observed herein may be as a result of high exposures due to work activity, social lifestyle/habit (eg, smoking) or unstable responses and coordination of the immune system.^38,39

The BMI has been investigated as a predictor of severity and mortality in COVID-19 patients. According to the World Health Organization’s weight classification, BMI ≥ 30 (computed weight (kg)/height (m²) is classified as obese. An earlier Saudi study revealed 97.8% proportion of a large cohort of COVID-19 patients had BMI ≥ 30⁴⁰ and another recorded obesity in 11.5% of their patients¹⁶ suggesting increases in the incidence rate among Saudi population due to sedentary lifestyles.^41–43 This remains a health concern since it has been pin-pointed as a risk factor for COVID-19, even in younger persons.^44,45 Obesity (BMI ≥30) was recorded in 51.3% of our patients but it was not a defining risk factor to either severity or mortality of the disease.

As observed, COVID-19 infection was high among non-Saudi nationals even though nationality did not contribute significantly to severity. Our finding closely relates to an earlier study that showed that about 80% of their COVID-19 patients were non-Saudi nationals.⁴⁰ High infection rates among non-nationals may be as a result of shared housing units increasing disease exposure.⁴⁶ This further emphasizes the need for government regulations to reduce residences that are overpopulated thereby reducing the spread of COVID-19 infection and the burden of the pandemic.

The general mortality rate was 11.15% which is lower than the proportion (17.5%) reported in a previous Saudi study,⁴⁷ and could reflect a decline in agreement with a recent epidemiological study focused on Saudi population.³³ Mortality rates in some regions of the Gulf nations are reportedly higher; 37.5% for Sudan,¹⁹ 31.7% in North Darfur, 60% in Central Darfur,⁴⁸ which could be due to the lack of appropriate health care; testing and treatment facilities.⁴⁹ Lower mortality rates have, however, been recently reported in India (2–3%), Italy (0.07%), and the United States of America (6%).¹⁹

Of all our COVID-19 cases, 101 patients required ICU admission (17.88%) while earlier studies have reported 4.8%,⁵⁰ 4.7%¹⁸ and 50%.⁵¹ An interesting observation is that none of the patient’s characteristics evaluated in this study presented any significant risk to increasing the need of the patients for intensive care. Over half of the COVID-19 patients investigated herein exhibited various comorbidities, more than reported in previous studies⁵⁰ indicating increasing susceptibility among those with pre-existing conditions. In this study, we found that 46.9% of the patients had respiratory diseases, which was the highest comorbidity recorded, followed by diabetes (41.6%) and hypertension (33.8%). Other highly occurring comorbidity include cardiovascular (17.0%), renal (14.9%) and neurological (7.3%) diseases. This is in contrast to many Saudi and non-Saudi studies which showed diabetes and/or hypertension as the most common comorbidities.^{16,18,34,52–55} However, contrary to some earlier Saudi and non-Saudi studies,^{40,47,56–58} our findings revealed that hypertension and diabetes had no significant effect on the admission of the COVID-19 patients at the ICU, and as well do not affect the death rates. Several studies have also reported that no independent association exists between DM, hypertension and hospital in-death.^59,60 In addition, reduction in COVID-19 fatalities among diabetes patients as observed in this study may have been due to the rapt and attentive care administered to them by the health officials in the hospital since previous reports have already established them as high-risk patients. Moreover, our study identified that patients with respiratory [HR 4.0 (CI 2.0–8.1); p = 0.010], cardiovascular [HR₌4.1 (CI 2.5–6.7); p<0.001], and renal comorbidities were at significantly higher risk of COVID-19 severity, which requires admission at the ICU, eventually leading to death. This is in line with a previous study from the UK on a large group of patients the majority of which had chronic pulmonary and kidney diseases.¹³ Associations between these diseases and severe COVID-19 outcomes have also been reported in some non-Saudi studies.^61–64 A recent study from the Eastern Saudi province also revealed G6PD deficiency, which may lead to hemolytic anemia, as one of the most prominent comorbidity.⁵⁰ Reportedly, SARS-CoV-2 has negatively influenced the outcome of anemic patients by interacting with hemoglobin either by attacking heme group on 1-beta chain leading to hemolysis or by hepcidin-mimetic action of a spike protein causing hyperferritinemia.^65,66 Cardiac and chronic respiratory diseases have also been identified as risk factors associated with worse outcomes in COVID-19 patients⁶⁷ while lung and chronic kidney diseases were also recorded as risk factors in an independent study.⁶⁸ We also observed that neurological comorbidities contribute significantly to mortality among the hospitalized patients in line with some previous studies which reported cerebrovascular diseases leads to severe outcomes in COVID-19 cases.^64,69,70 However, after adjustments for all variables, admission of COVID-19 patients at ICU [HR 4.2 (CI 2.5–7.2); p<0.001] predicates mortality among the patients while renal and neurological comorbidities were not associated with fatalities.

Limitations

Although our study results succinctly examined various variables with regards to risk factors and clinical outcomes, it has some limitations which we acknowledge. Most importantly, the study population was curated from a single hospital in Riyadh, which is a more multicultural region compared to the rest of KSA. Therefore, findings herein cannot be generalized for all COVID-19 infected persons in the entire Kingdom.

Conclusion

Our study employed a large cohort of hospitalized COVID-19 patients in KSA and examined factors that could contribute to increases in disease severity and deaths. Findings revealed that incidences occurred across all the age-groups and were higher in males compared to females, and in non-Saudi nationals. However, this factors together with the BMI neither significantly determined patients’ admittance at the ICU or death. Some commonly reported comorbidities like hypertension and diabetes had no significant influence on fatalities among the patients. On the contrary, cardiovascular, respiratory and renal comorbidities were high-risk factors to worsening outcomes and even death among hospitalized patients. We believe our study results could facilitate health-care providers in identifying patients with risk of fatality. Findings will also be helpful to decision makers to continue deriving and implementing regulations to reduce the spread of COVID-19 and other future infections among nationals and non-nationals living in KSA.