Incidence of New-Onset Hypertension Post–COVID-19: Comparison With Influenza

Hypertension is one of the most common comorbidities reported among patients with COVID-19.^1,2 The clinical course of COVID-19 has been well documented to be more severe in patients with preexisting hypertension, including higher rates of hospitalization, critical illness, and mortality compared with those with normotension.^3,4 Weakened immune system, activation of the renin-angiotensin-aldosterone system (RAAS), and proinflammatory cytokines in patients with chronic hypertension may be contributing factors.⁵ Conversely, COVID-19 may trigger new-onset hypertension or exacerbate preexisting hypertension.⁶ Heart muscle has a high density of ACE2 (angiotensin-converting enzyme 2) receptors through which SARS-CoV-2 virus gains entry to infect host cells,⁷ which may lead to blood pressure dysregulation. Other mechanisms in the host response to COVID-19 that may lead to the development of hypertension include severe inflammation and increased cytokine release.^2,8–11 Renin-angiotensin system blockers upregulate ACE2, and there was concern that angiotensin-converting inhibitors and angiotensin receptor blockers may lead to increased severity of COVID-19 and mortality early in the pandemic, although this concern was subsequently refuted.^12–16

Although several case reports and commentaries have drawn attention to a possible association between COVID-19 and new-onset hypertension,^6,17,18 there have been no large cohort studies investigating this association. It is also unknown whether new-onset hypertension diagnosed during COVID-19 hospitalization persists after resolution of the acute infection. The effects of COVID-19-related complications on the heart and blood pressure regulation may resolve or there could be long-lasting effects on the cardiovascular system and RAAS memory after resolution of acute infection. Furthermore, it is unclear if the incidence of hypertension after COVID-19 differs from other respiratory viral infections such as influenza. Comparison with an appropriate control group is important because new-onset hypertension may be a natural occurrence with time as patients with COVID-19 are generally older and have preexisting comorbidities which place them at increased risk of developing hypertension.

The goals of this study were to determine the incidence of new-onset persistent hypertension following COVID-19, risk factors associated with new-onset persistent hypertension in patients with COVID-19, and whether the incidence was higher than the incidence following infection with influenza, a similar respiratory virus. We also compared the incidence of hypertension across the pandemic and with respect to COVID-19 hospitalization status. Predictive models were used to identify risk factors associated with new-onset persistent hypertension.

METHODS

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Data Sources

This study was approved by the Einstein-Montefiore Institutional Review Board (#2021-13658) with an exemption for informed consent. Health data came from the Montefiore Health System with 15 hospitals located in New York Metropolitan area in the Bronx and its environs, serving a large diverse patient population, including many patients with lower social economic status. Electronic medical records were extracted automatically and cross-validation of all major variables extracted by manual chart reviews on subsets of patients was performed to ensure data accuracy.^19–24 From March 1, 2020 to August 17, 2022, there were 45 398 unique COVID-19 positive patients, identified by polymerase-chain-reaction test. From Jan 2018 to Feb 20, 2022, there were 13 864 unique patients who tested positive for influenza without a positive COVID-19 polymerase-chain-reaction test. Influenza patient demographics pre (2018 and 2019) and post (2020–2022) pandemic were similar and were thus combined. Patients who had SARS-CoV-2 and influenza co-infection were excluded (N=1139, 2.5%).

Variables

Demographic data included age, sex, race, and ethnicity. Preexisting comorbidities included congestive heart failure, chronic kidney disease (CKD), coronary artery disease (CAD), diabetes, chronic obstructive pulmonary disease (COPD), and asthma that were designated by International Classification of Disease-Tenth Revision (ICD-10) codes at admission or prior. Hospitalization status, treatment with vasopressor and corticosteroid medications during hospitalization, intensive care unit admission, and all-cause mortality were also extracted. Admission vital signs and laboratory data collected from hospitalized patients included systolic blood pressure (SBP), diastolic blood pressure (DBP), O₂ saturation, respiratory rate, heart rate, LDH (lactate dehydrogenase), cholesterol, BNP (brain natriuretic peptide), creatinine, estimated glomerular filtration rate, blood urea nitrogen, high-sensitivity CRP (C-reactive protein), ferritin, D-dimer, troponin-T, hemoglobin, ALT (alanine aminotransferase), AST (aspartate aminotransferase), lymphocyte counts, leukocyte counts, monocyte counts, potassium, sodium, prothrombin time, albumin, along with baseline body mass index.

Time Points

Data were collected at admission and at follow-up. The follow-up was within 3 and 9 months after COVID-19 or influenza positive test, with the one closest to 6 months taken for both COVID-19 and influenza patients.

Definitions

History of Hypertension

Patients with a history of hypertension were excluded. History of hypertension was defined as those who met at least one of the 3 criteria pre–COVID-19: (1) patients had average blood pressure 2 weeks before COVID-19 diagnosis above 140 mm Hg systolic or 90 mm Hg diastolic blood pressure (please see sensitivity analysis using 130/80 mm Hg cutoffs), (2) patients had a previous ICD-10 hypertension code diagnosis before COVID-19 diagnosis, or (3) patients was prescribed at least one antihypertensive medication at the time of COVID-19 diagnosis (see appendix for ICD-10 codes and antihypertensive medications).

Persistent Hypertension

Persistent hypertension, the primary outcome, was defined as a patient who returned during the follow-up period who had met at least one of the 3 criteria: (1) minimum blood pressure during this visit was above 140 mm Hg systolic or 90 mm Hg diastolic, (2) patients had a new ICD-10 hypertension code diagnosis up to moment of follow-up, (3) patients were prescribed an antihypertensive medication at the follow-up visit.

In-Hospital Hypertension

In-hospital hypertension, the secondary outcome, was defined as those who met at least one of the 3 criteria during hospitalization: (1) patients for those who had >3 blood pressure measurements during hospitalization, the average blood pressure was either above 140 mm Hg systolic or 90 mm Hg diastolic, (2) patients had a new ICD-10 hypertension code diagnosis during hospitalization, or (3) patients were prescribed antihypertensive medications during hospitalization.

Sensitivity Analysis

We also analyzed incidence of persistent hypertension using blood pressure cutoffs of 130/80 mm Hg (in addition to 140/90 mm Hg cutoff).

Odds Ratios and Predictive Models

Logistic regression was used to identify top predictors of new-onset persistent hypertension for (1) hospitalized patients with COVID-19 using demographics, comorbidities, treatment (intensive care unit, vasopressors) and laboratory test data, (2) nonhospitalized patients with COVID-19 using demographics, comorbidities, and (3) hospitalized and nonhospitalized patients with COVID-19. Odds ratios were calculated. Univariate analysis was used to rank the top predictors of persistent hypertension by P value.

Prediction of persistent hypertension used all variables shown. Prediction was made with 80% of data used for training and 20% testing, along with 5-fold cross-validation predictors. Prediction model performance was evaluated using receiver-operating-characteristic area under the curve (AUC). Influenza data were not included in the model because very few patients with influenza developed new-onset persistent hypertension. There were essentially no missing demographic data and comorbidities data. Laboratory variables with >15% missing variables were excluded in prediction models.

Statistical Analysis

Statistical analysis was performed using Python and Statistical Analysis System software (Version 9.4). Group comparison for categorical variables used χ² exact tests, and for continuous variables used the Mann-Whitney U test. Adjusted odds ratio was calculated with age, sex, intensive care unit status, and admission CRP, LDH, and D-Dimer as surrogates for severity as covariates. Prediction model performance was evaluated using receiver-operating-characteristic AUC. P<0.05 was considered statistically significant unless noted otherwise.

RESULTS

Incidence of Persistent Hypertension

Figure 1 shows the patient selection flowchart. There were 45 398 patients with COVID-19 and 28 576 did not have a history of hypertension. Of the 5562 hospitalized patients with COVID-19, 1455 patients returned, and 298 (20.6%) developed persistent hypertension. Of the 23 014 nonhospitalized patients with COVID-19, 5565 returned and only 604 (10.9%) developed persistent hypertension.

• Download figure
• Download PowerPoint

There were 13 864 influenza patients without COVID-19, and 11 516 did not have a history of hypertension. Of the 619 hospitalized influenza patients, 147 returned, and 24 (16.3%) developed persistent hypertension. Of the 10 897 nonhospitalized influenza patients, 2400 returned and only 105 (4.4%) developed persistent hypertension.

Characteristics of Patients With Persistent Hypertension

Hospitalized patients with COVID-19 with persistent hypertension (Table 1) were older, more likely male, Black race (but no differences in ethnicity), have higher prevalence of congestive heart failure and CKD but lower prevalence of COPD, higher SBP, DBP, creatinine, blood urea nitrogen, and CRP but lower heart rate and O₂ saturation at admission compared with those without persistent hypertension (all P<0.05). They were also more likely to be given corticosteroids and vasopressors during COVID-19. Nonhospitalized patients with COVID-19 with persistent hypertension (Table 2) were older, more male, and more likely to be White or Black and less likely to be Hispanic (all P<0.05).

Hospitalized influenza patients with persistent hypertension (Table 1) were older, had higher SBP, DBP, bilirubin, and lower respiration rate at admission compared with those without persistent hypertension. Nonhospitalized influenza patients with persistent hypertension (Table 2) were older and had higher prevalence of diabetes and lower prevalence of COPD compared with those without persistent hypertension.

Comparison of Persistent Hypertension in COVID-19 Versus Influenza

Hospitalized (adjusted odds ratio, 2.23 [1.48–3.54]; P<0.001) and nonhospitalized COVID-19 (adjusted odds ratio, 1.52 [1.22–1.90]; P<0.01) patients showed higher persistent hypertension incidence compared with influenza counterparts. Demographics and comorbidities between COVID-19 and influenza who developed persistent hypertension were not significantly different between groups, except COPD, heart rate, albumin, and estimated glomerular filtration rate (as denoted by ‡ in Tables 1 and 2).

Risk Factors for Persistent Hypertension

For the hospitalized COVID-19 cohort, the most significant risk factors for developing persistent hypertension were age, Black race, SBP, COPD, CAD, and corticosteroid administration (Table 3). None of the laboratory data variables were top predictors. The model predicted new-onset persistent hypertension with 79.37±2.60% accuracy and 72.56±1.33% AUC. For the nonhospitalized COVID-19 cohort, the most significant risk factors for developing persistent hypertension were age, CKD, Black race, and male sex (Table 2). The model predicted new-onset persistent hypertension with 86.06±2.15% accuracy and 77.38±2.59% AUC.

When both nonhospitalized and hospitalized patients with COVID-19 were included in the model (Table 3), the most significant risk factors for developing persistent hypertension were age, Black race, CKD, COPD, and CAD. The model predicted new-onset persistent hypertension with 82.80±1.28% accuracy and 71.95±3.24% AUC.

Incidence of Persistent Hypertension Across the COVID-19 Pandemic

To distinguish the importance of disease severity in contributing to persistent hypertension, we analyzed data across the pandemic. COVID-19 positive cases showed 4 distinct waves (predominant viral strains) across the pandemic for both hospitalized (Figure 2A) and nonhospitalized cohort (Figure 2B). Incidence of persistent hypertension in the hospitalized cohorts were twice as high as that in nonhospitalized cohort (≈20% versus ≈10%). The hospitalized cohort showed no apparent trends of persistent hypertension incidence across the pandemic (Figure 2C, linear regression, R²=0.055, P=0.39). The nonhospitalized cohort showed persistent hypertension incidence decreased from ≈20% in March 2020 to ≈10% in October 2020 (R²=0.79, P=0.003) and then plateaued. Small samples sizes, however, precluded quantitative statistical comparison of persistent hypertension incidence for individual waves.

• Download figure
• Download PowerPoint

In-Hospital Hypertension

In a secondary analysis, we analyzed the incidence of in-hospital hypertension. Of the 5562 hospitalized patients with COVID-19 without history of hypertension, 1167 (21.0%) developed in-hospital hypertension. Of the 619 hospitalized influenza patients without history of hypertension, 85 (13.7%) developed in-hospital hypertension.

COVID-19 hospitalized patients with in-hospital hypertension were older, more likely to be male, more likely to be Black or White race, had higher prevalence of congestive heart failure, CKD but lower prevalence of COPD, had higher values of SBP, DBP, creatinine, blood urea nitrogen, potassium, and CRP, but lower bilirubin, heart rate, albumin, respirate rate, and leukocytes at admission compared with those without in-hospital hypertension (all P<0.05, Table 4). They were more likely to be treated vasopressors (P<0.001) and corticosteroids (P<0.001).

Influenza hospitalized patients with in-hospital hypertension were older, higher prevalence of congestive heart failure, had higher SBP, DBP, hemoglobin, creatinine, troponin, blood urea nitrogen, and body mass index, but lower heart rate and respiration rate at admission compared with those without in-hospital hypertension (all P<0.05, Table 3). They were more likely to be admitted to intensive care unit (P<0.05) but similar mortality rate (P>0.05).

Sensitivity Analysis

As a sensitivity analysis, we analyzed the data using SBP/DBP threshold of 130/80 mm Hg (in addition to 140/90 mm Hg; Table 5). There were 45 398 patients with COVID-19 and 23 650 who did not have a history of hypertension. Of the 4117 hospitalized patients with COVID-19, 1054 patients returned, and 367 (34.8%) developed persistent hypertension. Of the 19 533 nonhospitalized patients with COVID-19, 4553 returned and 1172 (25.7%) developed persistent hypertension. The general observations with respective to differences in demographics, comorbidities, laboratory data and outcomes were similar as those reported for 140/90 mm Hg cutoff for both hospitalized and nonhospitalized patients with COVID-19.

There were 13 864 influenza patients without COVID-19, and 10 024 did not have a history of hypertension. Of the 519 hospitalized influenza patients, 119 returned, and 23 (19.3%) developed persistent hypertension. Of the 9505 nonhospitalized influenza patients, 2099 returned and only 191 (9.1%) developed persistent hypertension. These are shown in Table 6. The overall findings were similar with 140/90 mm Hg cutoffs, with some outcomes showing higher incidence.

To assess potential anxiety effects on blood pressure, we also performed sensitivity analysis using 2 separate visits during follow-up and defined the criteria of persistent hypertension using the 140/90 mm Hg cutoff for both. Two separate blood pressure measurements greater than 140/90 mm Hg from 2 visits within the follow-up period of 3- to 9 months were used. Note that the minimum blood pressure of each visit was used, which further accounts for possible elevations due to white-coat syndrome or anxiety. Of the 5562 hospitalized patients with COVID-19, 857 returned at 2 separate visits during the follow-up period with blood pressure measurements, and 151 (17.6%) developed persistent hypertension (Table S1A). Of the 23 014 nonhospitalized patients with COVID-19, 2670 returned at 2 separate visits with blood pressure measurements, and 186 (7.0%) developed persistent hypertension (Table S1B).

Of the 619 hospitalized influenza patients without a history of hypertension, 96 returned with 2 separate visits including blood pressure measurements, and 11 (4.4%) developed persistent hypertension (Table S1C). Of the 10 897 nonhospitalized influenza patients, 840 returned 2 separate times with blood pressure measurements, of which 40 (4.8%) developed persistent hypertension (Table S1D). The overall findings suggested slightly lower incidence of persistent hypertension.

DISCUSSION

This study investigated the incidence of new-onset hypertension associated with COVID-19 compared with influenza in a diverse population in a large health care system in Bronx, New York. The major findings are (1) new-onset persistent hypertension at follow-up is seen in 20.6% of hospitalized patients with COVID-19 and 10.85% of nonhospitalized patients with COVID-19 with no prior history of hypertension, (2) hospitalized patients with COVID-19 are 2.23 times and nonhospitalized patients with COVID-19 are 1.52 times more likely to develop new-onset persistent hypertension than their influenza counterparts, (3) patients of older age, Black race, male sex, and major comorbidities were at higher risk for persistent hypertension, and (4) there were 21.0% of hospitalized patients with COVID-19 with no prior history of hypertension who developed in-hospital hypertension and risk factors were similar to those who developed persistent hypertension.

Possible Mechanisms

To our knowledge this is the first cohort study identifying the association between new-onset persistent hypertension and COVID-19 in patients with no prior history of hypertension. The mechanisms underlying how SARS-CoV-2 triggers new-onset hypertension are unknown. It is possible that SARS-CoV-2 causes cardiac dysfunction and blood pressure dysregulation as there is evidence that SARS-CoV-2 directly infects cardiac cells via the ACE2 receptors.⁷ The RAAS plays an important role in blood pressure regulation and COVID-19 disease may stimulate the RAAS. Persistent activation of RAAS and endothelial injury have been reported among patients with COVID-19 and both are associated with blood pressure elevation.^25,26 In addition, consequences of severe COVID-19 including systemic hypoxia, acute respiratory distress, hypercoagulation, sepsis, inflammation, metabolic stress, and cytokine storm may stress the cardiovascular system that could lead to blood pressure dysregulation.^2,8–11

Acute kidney injury (AKI), a common complication among hospitalized patients with COVID-19,^3,4,27,28 may also lead to new-onset hypertension secondary to fluid retention.²⁹ A previous study of >43 000 hospitalized patients found that AKI survivors had 22% higher odds of developing hypertension compared with those without AKI. AKI has been linked to later development of cardiovascular events and new-onset hypertension or worsening blood pressure among those with a history of hypertension may be contributing mechanisms.³⁰ Consistent with these findings, we also found higher AKI incidence among those with new-onset hypertension compared with normotensive patients. Previous studies have shown that there is an inverse relationship between the prevalence of hypertension and worsening stage of CKD.³¹ Consistent with this observation, CKD was a top predictor of new-onset hypertension in our models.

Finally, other factors contributing to the development of hypertension following COVID-19 may include the effects of isolation, psychosocial stress, reduced physical activity, unhealthy diet, and weight gain during the pandemic. A recent AHA study reported found a significant rise in blood pressure (1–3 mm Hg, depending on age) among US adults during the COVID-19 pandemic compared with the corresponding periods before pandemic.³² Small population level increases in blood pressure could increase long-term incidence of major adverse cardiovascular events (ie, a 2-mm Hg higher SBP is associated with significant increases in mortality from stroke and ischemic heart disease among middle-aged adults).³³

Risk Factors

Patients of older age, male sex, and with major comorbidities (COPD, CAD, CKD) are at higher risk of persistent hypertension. Many prior COVID-19 studies have reported that male sex is associated with worse acute outcomes, including multiorgan injury, critical illness, and mortality.^3,4,21,28,34 Sex differences in susceptibility include differential expression and regulation of ACE2, innate inflammatory response, and adaptive immunity.³⁵ COPD, CAD, and CKD are known risk factors for hypertension.³⁶ Interestingly, diabetes status was not a top risk factor in our study. No laboratory test variables or vital signs (except systolic blood pressure) in the hospitalized cohorts were found to be top risk factors. Prediction accuracy for persistent hypertension was high using common clinical data. Hypertension is strongly associated with adverse cardiovascular outcomes and mortality. This prediction model may be useful to identify patients at risk of developing hypertension following COVID-19 and thus could have important clinical utility, as it would allow for earlier initiation of treatment and assessment for hypertension related complications such as cardiovascular and kidney disease. Influenza data were not included in the model because of small sample size (comparatively very few patients with influenza developed significant in-hospital and persistent hypertension).

Incidence of Persistent Hypertension Across the Pandemic

There was no apparent trend of persistent hypertension incidence across the pandemic in the hospitalized cohort, whereas persistent hypertension incidence significantly decreased across the pandemic in the nonhospitalized cohort from March 2020 to October 2020 and then plateaued thereafter. A possible explanation of the rapid drop in persistent hypertension incidence early in the pandemic is that patients had more severe COVID-19 illness, and effective treatments were not yet widely available. It should be noted that COVID-19 vaccine was only available until late Dec 2020 and thus this observation could not be explained by vaccine availability. The incidence of new-onset persistent hypertension in patients with COVID-19 remained high and time invariant after vaccine became widely available and higher than that in patients with influenza, and this is concerning.

Sensitivity Analysis

Sensitivity analysis showed that the overall outcomes and conclusions were not altered by blood pressure definition (140/90 versus 130/80 mm Hg). We also performed sensitivity analysis to assess if anxiety or white-coat syndrome affected our major outcomes. As expected, the return rate was comparatively lower. Note that the sample size of hospitalized influenza patients was small (N=96), and significant differences need to be interpreted with caution. Nonetheless, the major outcomes did not seem to be significantly affected by anxiety or white-coat syndrome.

Comparison With Influenza

We quantitatively compared new-onset persistent hypertension of COVID-19 with that of influenza, a similar respiratory virus, in the same catchment area. A major finding is that the incidence of new-onset persistent hypertension after COVID-19 was markedly higher than that after influenza infection. Influenza patients were younger overall and had comparatively few comorbidities, except COPD.

There are common risk factors among the COVID-19 and influenza cohorts. The increased risk of persistent hypertension associated with COVID-19 appears to be mediated through preexisting medical conditions and disease severity. The high prevalence of patients infected with COVID-19 compared with influenza suggest that many more patients will develop hypertension following COVID-19 in the population.

Limitations

There are several limitations in our analysis. Our findings were restricted to patients who returned to our health system. It is possible that patients who returned were more likely to have more severe COVID-19. However, patient data obtained via electronic medical records included those who returned for any medical reason, including but not limited to routine office visits and blood pressure measurements when available. A subanalysis comparing the characteristics of those who returned and those who did not showed similar demographics, comorbidities, and laboratory data at admission. It is also possible that some patients had undiagnosed hypertension, which could result in misclassification. Our cohort was minority predominant with low socioeconomic status which could increase susceptibility to developing hypertension in COVID-19 conditions. We did not compare to COVID-19-negative patients in general to avoid bias by hospitalized patients with COVID-19 who were likely admitted for serious medical issues (such as critical illness). Future studies should explore comparisons using propensity-matched COVID-19-negative controls.^37,38

The incidence of hypertension across the pandemic might be affected by other factors including the vaccination rate, strain of SARS-CoV-2, testing rate, population profile, and disease severity. Vaccine status was not reliably recorded if patients received vaccine outside our health care system. Vaccines and boosters were also administered in multiple stages based on age, and multiple doses, and types (some requiring 1 shot and others 2 shots) in the population. Thus, vaccine status is difficult to analyze with respect to our outcomes. COVID-19 testing rate and the profile of the patients (ie, more older patients were affected earlier in the pandemic) could also affect the temporal incidence of new-onset hypertension. The effects of these confounders on outcomes are complex, difficult to assess, and not readily discernable from one another. We followed patients for ≈6 months after diagnosis, but a longer follow-up study would be needed. Time to event survival analysis could be used to gain further insights into the evaluation of new-onset hypertension As with any retrospective study, there could be other unintended patient selection bias and latent confounding.^39,40

Conclusions

The incidence of new-onset persistent hypertension in patients with COVID-19 is much higher than in patients with influenza. The high incidence is not limited to people with severe COVID-19 illness. This statistic is alarming given the mass number of people affected by COVID-19. New-onset persistent hypertension may constitute a major cardiovascular sequela at a population level. Identifying risk factors for developing new-onset hypertension may draw clinical attention for the need for careful follow-up in at-risk individuals post–COVID-19 infection.

Perspectives

COVID-19 may trigger new-onset persistent hypertension. It is unclear whether this sequela differs from what occurs following other severe respiratory viral infections. We investigated the incidence and risk factors associated with new-onset persistent hypertension during COVID-19 and at 6-month follow-up compared with influenza. This work was performed in a major academic health system in the Bronx, serving a large racially and ethnically diverse population with a high proportion of patients with low socioeconomic status. We found that the incidence of new-onset persistent hypertension in patients with COVID-19 was significantly higher compared with patients with influenza. The mass number of people affected by COVID-19 suggests that new-onset persistent hypertension could be a major postinfection cardiovascular sequela at the population level. These findings should heighten awareness to screen for hypertension following COVID-19 illness in at-risk patients.

ARTICLE INFORMATION

Author Contributions

V. Zhang participated in concept and design, collected/verified data, analyzed data, created tables and figures, drafted article. W. Hou collected/verified data, validated data, statistical analysis. M. Fisher participated in concept and design, edited article. L. Zhang participated in concept and design, edited article. T.Q. Duong participated in concept and design, supervised, edited article.

Footnotes