Risk Factors and Blood Pressure Control Targets Associated With the Clinical Outcome of Elderly COVID-19 Patients With Hypertension: a Retrospective Cohort Study


 Background: COVID-19 is a global pandemic, especially among the elderly. Our study aimed to explore the risk factors and identify the blood pressure control targets associated with the clinical outcome of elderly COVID-19 patients with hypertension. Methods: In this retrospective cohort study, elderly COVID-19 patients who were admitted to Wuhan Huoshenshan Hospital from February 8 to 17, 2020 was included. Demographic, medical history, clinical data, and laboratory test data were collected from medical records. The adverse clinical outcomes were intensive care unit (ICU) admission and death. Difference between hypertension and non-hypertension groups were compared. Hypertension group were further divided into 3 subgroups according to their maximum blood pressures. Kaplan–Meier (K–M) method was used to find the differences both between hyperntesion and non-hypertension groups, and among the 3 hypertension subgroups. Univariable and multivariable Cox proportional hazards regression model were used to find risk factors.Results: All 133 elderly COVID-19 patients (79 patients with hypertension) were included. (1) Univariate analysis between hypertension and non-hypertension patients showed most laboratory tests were significantly (P < 0.05, or P < 0.01), particularly in adverse clinical outcomes (32.91% vs 7.41% at 30 days, P < 0.05). (2) Multivariate Cox proportional hazards models confirmed hypertension (HR 3.202, 95% CI:1.164 - 8.807) were the most important independent risk factors of outcomes in elderly patients, as well as low lymphocyte count, while the statistical difference of other values diminished. (3) Hypertension group were further divided into 3 subgroups according to their maximum blood pressures. K-M analysis showed maximum systolic blood pressure (SBP) ≥160mmHg subgroup (P < 0.01) and maximum blood pressure (DBP) ≥90mmHg subgroup (P < 0.05) experienced more adverse outcomes than others. (4) Multivariate Cox-proportional hazard model confirmed that maximum SBP≥160mmHg and maximum DBP ≥90mmHg were risk factors (HR 8.279, 95% CI: 1.346, 50.914; HR 5.080, 95% CI: 1.606,16.071; respectively). Conclusions: Hypertension is the most important independent risk factor of adverse outcomes in elderly COVID-19 patients, controlling the maximum blood pressure levels under 160/90 mmHg will decrease large part risks of adverse outcomes, the first week are key treatment period for patient prognosis.


Background
Coronavirus disease-2019 (COVID-19) is a life-threatening infection caused by the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) virus (1). It has caused billions infections and millions deaths (2,3). Based on the global pandemic of COVID-19, it has become critical to study the risk factors of clinical prognosis.
It has been previously reported that patients with older age and chronic underlying conditions might have a high risk of experiencing intensive care and death. Moreover, current clinical data suggest that, in COVID-19 patients, hypertension can worsen outcomes and increase the risk of admission to intensive care unit (ICU) (4). Previous clinical trials showed that SARSCoV-2 enters cells through angiotensinconverting enzyme 2(ACE2) receptors (5); high blood pressure is likely to increase COVID-19 mortality by affecting lung function and impairing oxygen delivery(6). As the COVID-19 pandemic progresses, the high incidence of hypertension makes this problem particularly worrying. However, it is not clear whether the mortality of COVID-19 patients with hypertension is related with the level of blood pressure control (7); there is a lack of detailed prognosis studies of hypertension at different blood pressure levels; even not to say few studies have explicitly identi ed blood pressure control cut-off points in hypertension comorbidities(8).
Thus, in this study, we aim to report the correlation between hypertension and COVID-19 adverse outcomes in elderly hypertensive population, and try to found the risk factors and the control cut-off value for elderly hypertension population.

Study design and participants
This single-center, retrospective cohort study was conducted at the Huoshenshan Hospital in Wuhan, which was approved by the Institutional Ethics Committee of Huoshenshan Hospital (HSSLL030).
Considering the urgency of the COVID-19 pandemic, the need for informed consent forms were waived by the ethics board of the hospitals. All participants included in the study were diagnosed with COVID-19 who were hospitalized at Huoshenshan Hospital from February 8 to 17, 2020. According to the World Health Organization's temporary guidelines, COVID-19 was diagnosed based on SARS-CoV-2 nucleic-acid swab test and/or chest computed tomography (CT) scanning (the criteria of the New Coronavirus transcription-polymerase chain reaction (RT-PCR) and/or gene diagnosis. And severe COVID-19 patients met the following criteria: (1) respiratory distress, with respiratory rate ≥ 30 beats/min, (2) nger oxygen saturation ≤ 93% at rest, and (3) ratio of the partial pressure of arterial oxygen to the fraction of inspired oxygen (PaO2:FiO2) ≤ 300 mmHg.
Hypertension was designated based on the patient's medical history and guideline for diagnosis according to their blood pressure level at admission(higher than 140/90mmHg). Patients with transient hormone-induced hypertension in the ICU were not included in the hypertension group. Blood pressure of patient were all measured and recorded several times depending on the patient's condition, the classi cation of hypertension made depending the 2020 International Society of Hypertension Global Hypertension Practice Guidelines (10). Maximum blood pressure, minimum blood pressure, pulse pressure and mean arterial pressure were all extracted based from the above data. Diabetes, cardiac injury, acute liver injury, and acute kidney injury status were designated based on the patient's medical history and guideline for diagnosis (11).

Statistical analysis
Our statistical analyses were conducted with SPSS (version 26.0) and GraphPad Prism (version 8.0) software. We made no assumptions regarding missing data. Categorical variables are described as frequency rates and percentage (%). Continuous variables are described as mean ± standard deviation (SD) or median (interquartile range (IQRs)). We used the Student's t-tests (normally distributed) or Mann Whitney test (non-normally distributed) for continuous variables between two groups, and used the Fisher's exact test or χ² test for categorical variables between two groups. To explore the risk factors associate with adverse clinical outcomes, univariable and multivariable Cox proportional hazards regression model was performed. All the variables included in the nal models were based on clinical and scienti c understanding, previous ndings, and the results of univariable analyses. We divided the hypertension group into different blood pressure subgroups according to the latest guideline and analyzed the composition ratio of each subgroup. Kaplan-Meier (K-M) method was further used to nd the best blood pressure control level in the COVID-19 hypertension group. A two-sided P value ,0.05 was considered statistically signi cant.

Population
A total of 180 patients were initially screened for the study, and 47 patients younger than 60 years old were excluded. According to the clinical diagnosis and/or medical history on admission, the 133 patients were divided into two groups: hypertension patients (n = 79) and non-hypertension patients (n = 54) ( Fig. 1). A total of 36 (27.1%) patients were de ned as adverse clinical outcomes, including ICU admission (n = 34, 25.6%) and in-hospital death (n = 16, 12.0%, of which 14 had overlapping with ICU patients).

The difference of clinical laboratory between hypertension and non-hypertension patients
Except c-reactive protein (CRP) and D-dimer, mean levels of other indices of the 133 patients were in normal range ( Table 2). All BG indices, kidney function (urea nitrogen, BUN), nutrition (albumin), myocardial enzyme (creatine kinase-MB, lactate dehydrogenase), in ammatory indices (CRP, white blood cell count, neutrophil percentage, neutrophil count, lymphocyte percentage), hemoglobin, coagulation indices (prothrombin time, PT) were signi cantly higher in hypertension patients than non-hypertension patients (P < 0.05). Other blood routine examination (lymphocyte count et al.) and biochemical detection (liver function et al.) in the hypertension group were inferior to the non-hypertension group (P < 0.05) ( Table 2).  Fig. 2A a-b). K-M analysis were performed to calculate the percent of clinical outcomes between hypertension group and non-hypertension group. At the last follow-up visit, 31 patients in the hypertension group (39.2%, in which 24.0% of on death in ICU and 15.2% of death) and 5 in the non-hypertension group (9.3%, in which 1.9% of on death in ICU, 7.4% of death) had adverse clinical outcomes ( Fig. 2A c-d), and the overall clinical outcomes rates in the two groups differed signi cantly on the basis of the strati ed log-rank test. It showed 32.91% (at 30 days) and 39.24% (at 60 days) adverse clinical outcomes in hypertension patients, compared to the 7.41% (at 30 days) and 9.26% (at 60 days) in non-hypertension patients. The adverse event in hypertension patients began to increase sharply at about 5 days and relaxed after 30 days. And it reached the platform at about 45 days (Fig. 2B).

Risk factors of adverse clinical outcomes in elderly COVID-19 patients
Univariate Cox hazard analysis was used to screen for statistically signi cant variables associated with adverse clinical outcomes. In the total population, hypertension (HR 4.937) and diabetes (HR 5.821) were the most signi cantly two risk factors related with adverse clinical outcomes. Average BG, BUN, neutrophil percentage, neutrophil count, maximum BG, white blood cell count, PT were the following 8 risk factors with HR range from 1.085-1.154 (Table 3). The Cox hazard analysis also showed that the top 10 risk factors of univariate Cox analysis in hypertension group were the same with that of total population only different in rank order (  The cut-off value of blood pressure related with adverse clinical outcomes in elderly hypertension COVID-19 patients The hypertension population were divided into 3 subgroups according to their blood pressure level. 40.5% had a maximum SBP of higher than 160mmHg; and 49.4% had a maximum DBP of higher than 90mmHg. (Fig. 4A).
Signi cant difference of clinical outcomes could be seen among different SBP subgroups in the hypertensive group. The difference of overall clinical outcomes was not statistically between the SBP (1) subgroup and the SBP (2) subgroup (P = 0.412 by the log-rank test), but was statistically between the SBP (1) subgroup and the SBP (3) subgroup (P = 0.002 by the log-rank test). There were 11.11% (at both 30 days and 60 days) cases with adverse clinical outcomes in SBP subgroup (1), 13.79% (at 30 days) and 20.69% (at 60 days) cases in SBP subgroup (2), 53.12% (at 30 days) and 71.87% (at 60 days) cases in SBP subgroup (3). At the same time, the difference of overall clinical outcomes between the DBP (1) subgroup and the DBP (2) subgroup (P = 0.005 by the log-rank test) or the DBP (3) subgroup (P = 0.02 by the log-rank test) was statistically. There were 17.50% (at 30 days) and 22.50% (at 60 days) patients with adverse clinical outcomes in the DBP subgroup (1), 47.83% (at 30 days) and 52.17% (at 60 days) patients in the DBP subgroup (2), 37.5% (at 30 days) and 62.5% (at 60 days) patients in the DBP subgroup (3).
The difference among maximum SBP and DBP subgroups all increased sharply at about 5 days and continued over time, the increase was most pronounced at 30 days, and reached the platform at about 45 days (Fig. 4B).
The results of multivariate Cox-proportional hazard model analysis showed that maximum SBP (3), maximum DBP (2) and maximum BG were the independent risk factors of adverse clinical outcomes in hypertension population (Fig. 4C).

Discussion
Our study provides the clinical characteristics of elderly concerning the role of comorbidities in the setting of COVID-19. Similar to the previously reported results, hypertension, diabetes, and cardiovascular diseases are the most frequent comorbidities (7,11,12). However, in our cohort, patients with hypertension reach 59.4% (79 of 133), 27.1% of total patients and 39.2% of the hypertension group have adverse clinical outcomes, which are higher than other studies (13). This might because the Huoshenshan Hospital prefers to treat severely ill patients and all enrolled cases are elderly in our study.
K-M analyze, univariate and multivariate Cox hazard analysis in the total patients con rmed that hypertension is the most independent risk factor of developing adverse clinical outcome in elderly COVID-19 patients, which consistent with previous research (14,15). The pathophysiologic mechanisms are generally believed to be the following two aspects: a critical interaction between SARS-CoV-2 and the angiotensin-converting enzyme 2(ACE2) (16, 17); the hypotension medicines which act on ACE2(18, 19).
Our study shows that the percentage of good outcomes in hypertension decreases 50% and the risk of adverse outcomes increases more than 3 times. COVID-19 patients with hypertension should be given additional attention to prevent worsening of their clinical outcomes (4,18). Instruction of blood pressure control in hypertensive patients is very important for clinical treatment of COVID-19.
We identify that among the many blood pressure indexes, the maximum systolic and diastolic blood pressure are independent risk factors. Further multivariate Cox-proportional hazard model indicates that maximum blood pressure ≥ 160/90 mmHg is associated with increased adverse clinical outcomes in hypertension group. Former studies also suggest that low and stable BP are optimal to achieve a favorable prognosis for COVID-19 patients with coexisting hypertension (19)(20)(21), but no exact treatment target was put forward. Only a Spanish study showed that admission SBP < 120 and ≥ 140mmHg was a predictor of higher all-cause mortality compared to SBP between 120-140mmHg(8). Our study shows a brand-new evidence for hypertension COVID-19 patients' treatment: it is no more be t to strictly control blood pressure lower than 160/90 mmHg to the completely normal range.
The other highlight of our study is the treatment time point in COVID-19 patients with hypertension. The good outcomes in higher hypertension group of ≥ 160/90 mmHg decrease sharply after the rst week, then reached about 50% at one month, 50-70% at the second month, the effect of SBP plays more important role than DBP. Thus, we believe that early blood pressure control in elderly COVID-19 cases with hypertension was important for the clinical prognosis of disease.
Our research reveals several other prognostic factors in elderly COVID-19 patients with hypertension. (1) In addition to blood pressure, the most effective factor in the progression of COVID-19 hypertension patients in this cohort was maximum BG value, which was out of our expectation. This may be because the maximum BG value presents the worst pancreatic conditions and makes infection di cult to control (11,22). (2)  patients. Thrombocytopenia was resulting from the consumption and/or the reduced production of platelets in damaged lungs. Anemia was not observed notably, but the decrease in hemoglobin was frequent(28, 29). (5) In this analysis, age was not included in the risk factors after adjustment for confounders, which might be because patients were all elderly, and the age span was relatively small.
Our study has the following limitations. First, the sample size of the study was small, especially in the non-hypertensive group. Second, due to the retrospective study design, the absence of ambulatory blood pressure monitoring results in all patients has limited the analysis of blood pressure control. Third, we did not have access to the data of patients before admission, so we were unable to evaluate the patient's previous blood pressure control. Fourth, patients admitted to Huoshenshan hospital were generally more serious, which made our study had an unavoidable selective bias.

Conclusions
In summary, hypertension is the most important independent risk factor of adverse outcomes in elderly COVID-19 patients, controlling the maximum blood pressure levels under 160/90 mmHg will decrease large part risks of adverse outcomes, the rst week are key treatment period for patient prognosis.  The multivariate Cox-proportional hazard model analysis to further certify whether variables affected patient clinical outcomes in the total population