Low diastolic blood pressure and adverse outcomes in inpatients with acute exacerbation of chronic obstructive pulmonary disease: A multicenter cohort study

Introduction

Chronic obstructive pulmonary disease (COPD) is a leading cause of morbidity and mortality worldwide that causes a substantial and increasing economic and social burden.[1] The global prevalence of COPD is estimated to be approximately 11.7%, with approximately three million deaths annually.[2] Acute exacerbation of COPD (AECOPD), a main cause of death in patients with COPD, always involves severe health conditions, a change in regular medication, and even adverse outcomes such as in-hospital mortality, invasive mechanical ventilation, and ICU admission.[3] With increased severity, the in-hospital mortality of AECOPD varies from 0.1% to 24.5%.[4–7] As a result, assessment of the severity and outcome prediction of patients with AECOPD at admission are vital in planning an appropriate treatment strategy.

Many researchers have studied the prognostic factors of adverse outcomes, which mainly included intensive care unit (ICU) admission, longer hospital stays, and short- or long-term mortality, in inpatients with AECOPD. Older age, laboratory tests indicating infection or acidosis, and cardiovascular and pulmonary comorbidities or severe dyspnea are known as predictive risk factors for adverse outcomes.[8–13] Nevertheless, the prognostic value of on-admission vital signs that can be easily obtained from inpatients with AECOPD, including blood pressure, radial pulse, pulse pressure, and respiratory rate, has rarely been studied.

Blood pressure management is an important issue among clinicians. Systolic blood pressure (SBP) garners greater attention than diastolic blood pressure (DBP), which is often overlooked in clinical practice. SBP is mostly discussed in natural populations or patients with cardiovascular disease (CVD);[14–16] but according to the literature, DBP is more frequently associated with prognosis in CVD and other diseases, and this association could be independent of SBP.[17–19] DBP has been acknowledged as an explicit and independent risk factor for the development of CVD in high-risk patients, although controversy remains.[20–22] In most studies, a J-curve or U-shaped relationship has been found between DBP and the risk of CVD occurrence in patients with a high risk of CVD, sometimes even in people with good health; that is, DBP >80 mmHg or DBP <60 mmHg usually predicts a greater risk of CVD.[21,23–26] High DBP alone has been associated with better outcomes in patients with congenital heart defects.[27] The association of DBP with mortality in patients with or at risk for CVD has also been intensively studied, although few conclusions have been reached. Some studies show a U-shaped relationship between mortality risk and DBP in some populations, including patients with COPD;[18,28–30] however, other studies have found that a similar relationship does not exist in patients with or without CVD.[24,31–35] Nevertheless, all the aforementioned studies indicate the vital prognostic function of low DBP for all-cause mortality among populations with or at increased risk of CVD.

However, the role of DBP in predicting the outcomes of inpatients with AECOPD has not been reported. Thus, this prospective multicenter study primarily aimed to evaluate the relationship between DBP on admission and in-hospital mortality, invasive mechanical ventilation, and ICU admission in patients hospitalized for AECOPD, with and without CVD.

Methods Ethical approval

The Ethics Committee of each participating academic medical center approved the study based on the ethical approval obtained from the Ethics Committee of the coordinating center, West China Hospital of Sichuan University (No: 2019-1056). All participants provided written informed consent.

Patients and study design

Patient inclusion was based on the prospective, non-interventional, multicenter cohort MAGNET AECOPD (MAnaGement aNd advErse ouTcomes in inpatients with acute exacerbation of COPD), a registry study in China (ChiCTR2100044625). The main aims of this registry study were to investigate the adverse outcomes (including intensive treatment usage; in-hospital venous thromboembolism; short- and long-term mortality; and readmission) of inpatients with AECOPD and to establish and validate early warning models of these adverse outcomes.[36] The original study enrolled consecutive adult inpatients diagnosed with AECOPD among 10 tertiary hospitals in China between September 2017 and July 2021. The diagnosis of AECOPD was based on the following criteria: (1) a history of COPD, including exposure to risk factors (e.g., tobacco smoking, specific environmental exposure); long-term dyspnea, chronic cough, or sputum production; and post-bronchodilator spirometry testing showing a forced expiratory volume in 1 s/forced vital capacity ratio <70%; and (2) an acute worsening of respiratory symptoms resulting in additional therapy. All the patients enrolled in the MAGNET AECOPD study were included in this analysis except for the patients with missing records of blood pressure on admission. Patients were managed at the discretion of their treating physicians.

Data collection

All physicians in charge of data collection were well trained. A standardized case report form was completed for every enrolled patient, including baseline demographics, comorbidities, vital signs on admission, laboratory results, and treatments. Blood pressure assessment was based on routine measurement in the medical center where the patient received treatment, and was completed within 2 h of admission using a wrist-type or brachial automatic sphygmomanometer, or mercurial sphygmomanometer. The values of SBP and DBP obtained by these methods were averaged over two measurements. For some patients admitted to the ICU, the first blood pressure records from invasive blood pressure monitoring were used. Patients were defined as having CVD if they had any of the following conditions: coronary heart disease, heart failure, arrhythmia, heart valve disease, or stoke, according to the American Heart Association.

Study outcomes

The primary outcome was all-cause in-hospital mortality. The secondary outcomes were invasive mechanical ventilation and ICU admission during the hospital stay.

Statistical analysis

All collected data were analyzed using R software version 4.1.3 (The R Project for Statistical Computing, Vienna, Austria) and IBM SPSS version 25.0 (IBM Corp., Armonk, NY, USA), and a P-value <0.05 was used to indicate statistically significant differences. Eligible patients were divided into a group with DBP <70 mmHg and a group with DBP ≥70 mmHg (based on most studies investigating the association between DBP and adverse outcomes in patients with CVD or at risk of CVD).[29,31,34,37] Baseline demographics, comorbidities, vital signs on admission, laboratory results, and outcome variables were compared between the two groups. Categorical data are expressed as number with percentage, and the chi-squared test was used for comparisons between groups. Normally distributed data are expressed as mean ± standard deviation, with Student's t-test used for comparisons; non-normally distributed data are expressed as median (Q1, Q3) and a non-parametric test (Mann–Whitney U test) was applied for comparisons between the two groups.

To explore the independent prognostic factors of in-hospital mortality in the overall cohort, univariable Cox regression was performed and factors with a P-value <0.05 were included in least absolute shrinkage and selection operator (LASSO) regression analysis, which was applied to filter and shrink appropriate variables using the "glmnet" R package to minimize the risk of overfitting. The penalty parameter (λ) for the model was determined in 100-fold cross-validation following the minimum criteria (i.e., the value of λ corresponding to the highest partial likelihood deviance, which is referred to as λ1se in the following description). The chosen variables in LASSO regression were further included in multivariable Cox regression to determine the independent prognostic factors associated with in-hospital mortality. A forest plot was drawn to show the variables included in the prognostic model based on this multivariable Cox regression (individuals with missing values of the included variables were excluded). We also conducted sensitivity analyses, excluding patients with pneumonia or sepsis. Similar strategies of statistical analysis were used for the secondary outcomes (invasive mechanical ventilation and ICU admission).

In addition to binary classification of DBP as <70 mmHg or ≥70 mmHg, DBP on admission was further categorized in 5-mmHg increments from <50 mmHg to ≥100 mmHg to identify the trend in the risk of adverse outcomes with a change in DBP in the overall cohort and the subgroups with or without CVD. According to the independent prognostic factors identified in the above analysis, multivariable Cox analysis was performed again to evaluate the risk of in-hospital mortality for different DBP categories. According to multivariable Cox analysis, the adjusted hazard ratio (HR) and 95% confidence interval (CI) for each category of DBP were calculated with reference to the DBP 75 to <80 mmHg, where the HR was considered 1.

In survival analysis, Kaplan–Meier analysis and the log-rank test were used to describe and compare mortality between patients with DBP <70 mmHg and DBP ≥70 mmHg in the overall population and in subgroups with or without CVD.

Results Characteristics of the study population

A total of 14,007 patients admitted for AECOPD were enrolled in the original registration study. In the present analysis, we excluded patients with missing DBP records (n = 374), and a total of 13,633 inpatients with AECOPD were included [Supplementary Figure 1,https://links.lww.com/CM9/B513]. Among them, 197 (1.45%) died during hospitalization, 453 (3.32%) required invasive mechanical ventilation during their hospital stay, and 1021 (7.49%) patients were admitted to the ICU. A total of 3794 patients (27.83%) had CVD comorbidity.

Characteristics of the included patients with AECOPD are described in Table 1, among which 3108 patients (22.80%) had DBP <70 mmHg on admission and 10,525 patients (77.20%) had DBP ≥70 mmHg. As expected, DBP <70 mmHg group patients tended to be older (74.42 ± 10.44 years vs. 71.76 ± 10.27 years, t = –12.87, P <0.001), with a higher proportion of female patients (24.68% [767/3108] vs. 20.49% [2157/10,525], χ2=24.93, P <0.001) than DBP ≥70 mmHg group. Additionally, there were significant differences in the prevalence of CVD, especially coronary heart disease (13.90% [432/3108] vs. 10.27% [1081/10,525], χ2 =32.03, P <0.001), heart failure (15.22% [473/3108] vs. 10.31% [1085/10,525], χ2=57.15, P <0.001), arrhythmia (11.49% [357/3108] vs. 7.76% [817/10,525], χ2=42.28, P <0.001), and stroke (7.85% [244/3108] vs. 5.84% [615/10,525], χ2=16.38, P <0.001), which were more frequently observed in patients with DBP <70 mmHg than in those DBP ≥70 mmHg. Other than CVDs comorbidities, bronchiectasis (19.50% [606/3108] vs. 17.12% [1802/10,525], χ2=9.32, P =0.002), interstitial lung disease (9.30% [289/3108] vs. 8.10% [853/10,525], χ2=4.46, P =0.035), chronic pulmonary heart disease (23.33% [725/3108] vs. 20.00% [2105/10,525], χ2=16.15, P <0.001), pneumonia (32.40% [1007/3108] vs. 23.81% [2506/10,525], χ2=92.57, P <0.001), sepsis (0.68% [21/3108] vs. 0.29% [31/10,525], χ2=9.17, P =0.002), reduced mobility (15.99% [497/3108] vs. 12.19% [1283/10,525], χ2=30.54, P <0.001), diabetes (15.54% [483/3108] vs. 12.26% [1290/10,525], χ2=22.87, P <0.001), chronic renal failure (5.57% [173/3108] vs. 3.15% [332/10,525], χ2=39.13, P <0.001), and anxiety or depression (2.25% [70/3108] vs. 1.53% [161/10,525], χ2=7.52, P =0.006) were more frequently in patients with DBP <70 mmHg compared with those with DBP ≥70 mmHg. Regarding vital signs on admission, patients with DBP <70 mmHg tended to have lower SBP (117 [105, 129] mmHg vs. 134 [125,147] mmHg, U = –44.54, P <0.001) and slower radial pulse (86 [77, 98] vs. 88 [80, 100], U = -6.60, P <0.001) as well as higher pulse pressure (54 [44, 66] mmHg vs. 51 [40, 63] mmHg, U = -9.51, P <0.001) than patients with DBP ≥70 mmHg. In terms of laboratory tests, patients with DBP <70 mmHg had lower red blood cell counts (4.10 [3.62, 4.58]×1012/L vs. 4.34 [3.89, 4.77] ×1012/L, U = -15.29, P <0.001), hemoglobin (Hb) (123 [108, 137] g/L vs. 130 [117, 142] g/L, U = -16.00, P <0.001), and platelet counts (189 [139, 248] ×109/L vs. 192 [146, 249]×109/L, U = -2.33, P =0.020), and higher neutrophil percentage (NEUT%) (75.90% [66.10%, 84.80%] vs. 75.00% [65.60%, 84.20%], U = -2.02, P =0.040) as well as slightly higher partial pressure of oxygen in arterial blood (83.70 [68.30, 108.00] mmHg vs. 82.50 [68.10, 105.40] mmHg, U = -2.00, P =0.046) than patients with DBP ≥70 mmHg. Compared with patients with DBP ≥70 mmHg, the rates of in-hospital mortality (3.35% [104/3108] vs. 0.88% [93/10,525], χ2=102.18, P <0.001), invasive mechanical ventilation (6.24% [194/3108] vs. 2.46% [259/10,525], χ2=106.79, P <0.001), and ICU admission (11.35% [352/3108] vs. 6.36% [669/10,525], χ2=85.52, P <0.001) were significantly higher in patients with DBP <70 mmHg, and the length of hospital stay was significantly longer (9 [7, 14] days vs. 9 [7, 13] days, U = -4.32, P <0.001) in patients with DBP <70 mmHg.

Table 1 - Basic characteristics of the included AECOPD patients. Characteristics

Total

(N = 13,633)

DBP < 70 mmHg

(N = 3108)

DBP70 mmHg

(N = 10,525)

Statistical values P-value Baseline characteristics Age (years) 72.36 ± 10.37 74.42 ± 10.44 71.76 ± 10.27 -12.87¶ <0.001 Female 2924 (21.45) 767 (24.68) 2157 (20.49) 24.93** <0.001 Current smoking 2735 (20.06) 604 (19.43) 2131 (20.25) 0.99** 0.320 BMI (kg/m2)* 21.3 (18.71–24.03) 20.89 (18.29–23.53) 21.48 (18.83–24.22) -5.56†† <0.001 BMI ≥ 30 kg/m2* 206 (2.77) 45 (2.41) 161 (2.89) 1.18** 0.277 Comorbidities CVDs 3794 (27.83) 1046 (33.66) 2748 (26.11) 68.02** <0.001 Coronary heart disease 1513 (11.10) 432 (13.90) 1081 (10.27) 32.03** <0.001 Heart failure 1558 (11.43) 473 (15.22) 1085 (10.31) 57.15** <0.001 Heart valve diseases 248 (1.82) 68 (2.19) 180 (1.71) 3.07** 0.080 Arrhythmia 1174 (8.61) 357 (11.49) 817 (7.76) 42.28** <0.001 Stroke 859 (6.30) 244 (7.85) 615 (5.84) 16.38** <0.001 Hypertension 4579 (33.59) 1016 (32.69) 3563 (33.85) 1.46** 0.228 Pulmonary tuberculosis 621 (4.56) 157 (5.05) 464 (4.41) 2.28** 0.131 Bronchiectasis 2408 (17.66) 606 (19.50) 1802 (17.12) 9.32** 0.002 ILD 1142 (8.38) 289 (9.30) 853 (8.10) 4.46** 0.035 Asthma 456 (3.34) 97 (3.12) 359 (3.41) 0.62** 0.430 OSAHS 85 (0.62) 15 (0.48) 70 (0.67) 1.29** 0.256 Chronic pulmonary heart disease 2830 (20.76) 725 (23.33) 2105 (20.00) 16.15** <0.001 Pneumonia 3513 (25.77) 1007 (32.40) 2506 (23.81) 92.57** <0.001 Active cancer 727 (5.33) 184 (5.92) 543 (5.16) 2.75** 0.097 Sepsis 52 (0.38) 21 (0.68) 31 (0.29) 9.17** 0.002 Reduced mobility 1780 (13.06) 497 (15.99) 1283 (12.19) 30.54** <0.001 Diabetes 1773 (13.01) 483 (15.54) 1290 (12.26) 22.87** <0.001 Chronic renal failure 505 (3.70) 173 (5.57) 332 (3.15) 39.13** <0.001 GERD 131 (0.96) 38 (1.22) 93 (0.88) 2.90** 0.089 Anxiety–depression 231 (1.69) 70 (2.25) 161 (1.53) 7.52** 0.006 Osteoporosis 341 (2.50) 92 (2.96) 249 (2.37) 3.48** 0.062 Vital Signs Systolic blood pressure (mmHg) 130 (120–145) 117 (105–129) 134 (125–147) -44.54†† <0.001 Diastolic blood pressure (mmHg) 79 (70–87) 64 (59–67) 83 (76–89) -84.88†† <0.001 Mean arterial pressure (mmHg) 97 (88–105) 82 (76–86) 100 (94–107) -76.41†† <0.001 Pulse pressure (mmHg) 52 (40–63) 54 (44–66) 51 (40–63) -9.51†† <0.001 Radial pulse (/min)† 88 (79–100) 86 (77–98) 88 (80–100) -6.60†† <0.001 Respiratory rate (/min)‡ 20 (20–22) 20 (20–22) 20 (20–22) -2.64†† 0.008 Laboratory tests Red blood cell (×1012/L)§ 4.29 (3.82–4.73) 4.10 (3.62–4.58) 4.34 (3.89–4.77) -15.29†† <0.001 Hemoglobin (g/L)§ 129 (115–141) 123 (108–137) 130 (117–142) -16.00†† <0.001 White blood cell (×109/L)§ 7.70 (5.90–10.18) 7.78 (5.76–10.40) 7.69 (5.90–10.10) -0.36†† 0.716 NEUT (%)§ 75.20 (65.70–84.30) 75.90 (66.10–84.80) 75.00 (65.60–84.20) -2.02†† 0.040 EOSR (%)§ 1.00 (0.10–2.60) 0.90 (0.10–2.50) 1.00 (0.10–2.60) -1.68†† 0.094 Platelet (×109/L)§ 192 (145–249) 189 (139–248) 192 (146–249) -2.33†† 0.020 pH|| 7.41 (7.37–7.44) 7.41 (7.37–7.44) 7.41 (7.37–7.44) -1.72†† 0.086 PaO2 (mmHg)|| 82.90 (68.1–106.0) 83.70 (68.30–108.00) 82.50 (68.10–105.40) -2.00†† 0.046 PaCO2 (mmHg)|| 42.70 (36.90–51.60) 42.30 (36.40–51.60) 42.80 (37.00–51.60) -1.31†† 0.190 Lactate (mmol/L)|| 1.50 (1.20–1.90) 1.50 (1.20–1.90) 1.50 (1.20–1.90) -1.27†† 0.203 Adverse Outcomes In-hospital mortality 197 (1.45) 104 (3.35) 93 (0.88) 102.18** <0.001 Invasive mechanical ventilation 453 (3.32) 194 (6.24) 259 (2.46) 106.79** <0.001 ICU admission 1021 (7.49) 352 (11.35) 669 (6.36) 85.52** <0.001 Length of Stay (days) 9 (7–14) 9 (7–14) 9 (7–13) -4.32†† <0.001

Data are presented as the number (%), mean ± standard deviation, or median (Q1, Q3). *Available in 7435 patients; †available in 13,600 patients; ‡available in 13,589 patients; §available in 11,425 patients; ||available in 8879 patients. ¶t values; **χ2 values; ††U values. AECOPD: Acute exacerbation of chronic obstructive pulmonary disease; BMI: Body mass index; CVDs: Cardiovascular diseases; DBP: Diastolic blood pressure; EOSR: Eosinophil percentage; GERD: Gastroesophageal reflux disease; ICU: Intensive care unit; ILD: Interstitial lung disease; NEUT: Neutrophil percentage; OSAHS: Obstructive sleep apnea-hypopnea syndrome; PaCO2: Partial pressure of carbon dioxide; PaO2:Partial pressure of oxygen.


Factors associated with in-hospital mortality in the overall cohort

A univariable Cox analysis was conducted based on the overall cohort to explore the factors related to in-hospital mortality [Supplementary Table 1,https://links.lww.com/CM9/B513], and those variables with P-value <0.05 were included in LASSO regression. After λ1se was determined [Supplementary Figure 2,https://links.lww.com/CM9/B513], the selected variables were further included in multivariable Cox regression analysis. A forest plot was drawn according to multivariable Cox regression [Figure 1], in which DBP <70 mmHg was an independent risk factor for in-hospital mortality (HR=2.16, 95% CI=[1.53–3.05], P <0.001), after adjusting for other prognostic factors. Other independent risk factors included heart failure (HR=1.51, 95% CI=[1.10–2.09], P =0.012), chronic pulmonary heart disease (HR=1.72, 95% CI=[1.25–2.36], P <0.001), pneumonia (HR=2.28, 95% CI = [1.63–3.17], P <0.001), interstitial lung disease (HR=1.76, 95% CI=[1.15–2.69], P =0.009), sepsis (HR=1.95, 95% CI = [1.03–3.69], P = 0.041), reduced mobility (HR=1.79, 95% CI=[1.29–2.48], P <0.001), older age (HR=1.41, 95% CI = [1.20–1.66], P <0.001), higher radial pulse (HR = 1.19, 95% CI=[1.05–1.36],

留言 (0)

沒有登入
gif