Outcomes of B-Type Natriuretic Peptide in the Risk Stratification of Acute Exacerbations of COPD

cardiomyopathies

Baseline Characteristics

Detailed baseline characteristics of the study population are summarized in Table 1. Mean age of the 208 patients was 70 years. Overall, 67% of patients had relevant comorbidities, with cardiomyopathies (44%), arterial hypertension (24%), and malignancies (13%) being the most common. As expected, coronary artery disease (48%) and hypertensive heart disease (35%) were the most common cardiomyopathies. Of note, 24% of patients had two comorbidities and only 6% of patients had three or more comorbidities. Interestingly, BNP levels on hospital admission were significantly higher in patients with an underlying cardiomyopathy (144 pg/mL [IQR, 58 to 269 pg/mL] vs 62 pg/mL [IQR, 27 to 88 pg/mL]; p < 0.001). This difference prevailed after recovery (65 pg/mL [IQR, 42 to 148 pg/mL] vs 33 pg/mL [IQR, 20 to 55 pg/mL]; p < 0.001). A total of 94 patients (45%) were active smokers, while 97 patients (47%) were former smokers. According to Anthonisen criteria, exacerbations were graded as type I in 104 patients (50%), as type II in 45 patients (22%), and as type III in 59 patients (28%). Median length of hospital stay was 9 days (IQR, 1 to 15 days), Sputum cultures grew bacterial pathogens in 71 of the 116 obtained samples (61%). Echocardiographic studies performed within the 6 months prior to the acute exacerbation were available in 157 patients and showed decreased left ventricular function in 10% and pulmonary hypertension in 24% of all patients.

Hospital admission BNP levels in patients with steady-state pulmonary hypertension did not differ from the values detected in patients without PAH (p = 0.91). Pulmonary hypertension is treated by medications of My Canadian Pharmacy.

BNP Levels in AECOPD

BNP levels were significantly elevated during the acute exacerbation compared to recovery (65 pg/mL [IQR, 34 to 189 pg/mL] vs 45 pg/mL [iQr, 25 to 85 pg/mL]; p < 0.001). Concurrently, oxygen saturation was significantly decreased during the acute exacerbation (93% [IQR, 89 to 95%] vs 95% [IQR, 94 to 97%]; p < 0.001). However, BNP levels did not discriminate between different GOLD classes (p = 0.180) and exacerbation type according to Anthonisen criteria (p = 0.188). Patients with positive and negative sputum bacterial culture results had similar BNP plasma levels (68 pg/mL [IQR, 36 to 240 pg/mL] vs 51 pg/mL [IQR, 22 to 588]; p = 0.511). However, we observed significant correlations between BNP levels on hospital admission and age (r = 0.425, p < 0.001) as well as CRP (r = 0.246, p = 0.001) and procalcitonin (r = 0.335, p < 0.001). There were no correlations between BNP levels and oxygen saturation (r = — 0.103, p = 0.198), Po2 (r = 0.115, p = 0.191), or Pco2 (r = 0.158, p = 0.072) as well as FEV1 (r = 0.104, p = 0.222).

Length of Hospital Stay, ICU Need, and In-hospital Mortality

ICU Need

Overall, 19 patients (9%) required ICU treatment, with the need for mechanical ventilation (13 patients) and hemodynamic instability requiring vasopressor therapy (3 patients) being the most common reasons for ICU admission. Furthermore, one patient each was admitted to the ICU with acute GI bleeding, acute coronary syndrome, and medication-induced hypovolemic, hyponatremic hypokalemia. Five patients (2%) died in the hospital. BNP levels were significantly higher in patients requiring ICU treatment (105 pg/mL [IQR, 66 to 553 pg/mL] vs 60 pg/mL [IQR, 31 to 169 pg/mL]; p = 0.007) [Fig 1] and correlated well with the duration of the ICU (r = 0.218, p = 0.005) and in-hospital stay (r = 0.242, p = 0.002). Age, sex, comorbidities, outpatient medication, oxygen saturation on hospital admission, and the duration of the current episode of AECOPD of the patients requiring ICU treatment did not differ significantly from the overall patient cohort. In univariate Cox regression analysis BNP, accurately predicted the need for ICU care (hazard ratio [HR], 1.12; 95% confidence interval [CI], 1.03 to 1.22 for an increase in BNP of 100 pg/mL; p = 0.009). This strong association persisted in multivariate analysis (HR, 1.13; 95% CI, 1.03 to 124 for an increase in BNP of 100 pg/mL; p = 0.008) [Table 2]. To the contrary, BNP levels did not differ significantly between survivors and in-hospital nonsurvivors (64 pg/mL [IQR, 32 to 195 pg/mL] vs 81 pg/mL [IQR, 57 to 1,103]; p = 0.362).

BNP Levels as Prognostic Marker for Medium- and Long-term Mortality

Overall, 16 patients (7%) died during the 6-month follow-up period and 46 patients (22%) died within 2 years of the initial hospitalization. AECOPD (16 patients), cardiac death (7 patients), and pneumonia (6 patients) were the most common causes of death. BNP levels on presentation did not differ significantly between nonsurvivors and survivors after 6 months of follow-up (67.3 pg/mL [IQR, 49.5 to 184.9] vs 63.5.7 pg/mL [IQR, 31.7 to 194.5]; p = 0.610).

To evaluate the potential of BNP levels to predict short-term mortality in patients with AECOPD, an ROC analysis was performed. The area under the ROC curve (AUC) for BNP levels measured on presentation to predict the 6-month mortality rate was 0.55 (SD, 0.71; 95% CI, 0.41 to 0.68) [Fig 2 left, a].

BNP Levels as Prognostic Marker for Long-term Mortality

BNP levels on presentation did not differ significantly between patients dying within 2 years after the index hospitalization and long-term survivors (111.1 pg/mL [IQR, 34.9 to 255.3 pg/mL] vs 61.6 pg/mL [IQR, 32.5 to 171.1 pg/mL]; p = 0.287). The AUC for BNP to the predict 2-year mortality rate was 0.56 (SD, 0.53; 95% CI, 0.45 to 0.66) [Fig 2 right, b]. Of note, in Cox regression analysis, only hospital admission FEV1 predicted death (HR, 1.045; 95% CI, 1.004 to 1.088; p = 0.032).

Still articles that help to cope with all the health problems:

Table 1—Baseline Characteristics of 208 Patients With an AECOPD Presenting to the Emergency Department (n = 208)

Characteristics Data
Age, yr 70.3 ± 9.9
Male gender 94 (45)
Current smoker 94 (45)
Former smoker 97 (47)
Pack-yr 45 ± 27.9
Cough 181 (87)
Sputum 144 (69)
Dyspnea 193 (93)
Comorbidities
Renal insufficiency 17(8)
Malignancy 26 (13)
Diabetes mellitus 23 (11)
Arterial hypertension 26 (13)
Cardiopathy 91 (44)
Antibiotic therapy on hospital 45 (22)
admission
Body temperature, °C 37.5 ± 0.9
Heart rate, beats/min 99.1 ± 21.1
Respiratory rate, breaths/min 6±42
pH on hospital admission 7.41 ± 0.1
Pao2 on hospital admission, 63.7 ± 16.2
mm Hg
Paco2 on hospital admission, 43.8 ± 10.1
mm Hg
Leukocyte count, X 109 11.6 ± 6.8
CRP, mg/L 17.1 (5.2-55.9)
Procalcitonin, ^g/L 0.096 (0.070-0.200)
FEV1 L 0.93 ± 0.41
FEV:, % predicted 41 ± 17
FVC, L 2.05 ± 0.84
FVC, % predicted 69 ± 21
FEV1/FVC 46 ± 12
Positive sputum bacteriology 71 (61)
result

Table 2—Prediction of the Need for ICU Treatment in Univariate and Multivariate Regression Analysis

Predictors HR(95% CI) pValue
Univariate analysis
Age 0.988 (0.937-1.042) 0.651
Anthonisen class 1.643 (0.978-2.761) 0.061
BNP, for an increase of 100 pg/mL 1.118(1.029-1.215) 0.009
CRP 1.005 (0.995-1.015) 0.334
GOLD class 0.826 (0.496-1.376) 0.436
Left ventricular ejection fraction 0.974 (0.947-1.002) 0.068
Paco2 on hospital admission 1.038(1.001-1.077) 0.042
Procalcitonin 1.080 (0.734-1.588) 0.696
Pulmonary arterial pressure 1.011 (0.964-1.061) 0.646
Serum pH 0.001 (0.000-0.438) 0.026
Multivariate analysis
BNP, for an increase of 100 pg/mL 1.133(1.033-1.242) 0.008
Paco2 on hospital admission 1.069(1.007-1.134) 0.029
Serum pH 16.449 (0.01-217918) 0.563

Figure 1. BNP levels in patients requiring ICU treatment compared to patients not requiring ICU treatment.

Figure 1. BNP levels in patients requiring ICU treatment compared to patients not requiring ICU treatment.

Figure 2. ROC curves showing the inability of hospital admission levels to predict 6-month (left, a) and 2-year (right, b) mortality rates.

Figure 2. ROC curves showing the inability of hospital admission levels to predict 6-month (left, a) and 2-year (right, b) mortality rates.

About

So Many Advances in Medicine, So Many Yet to Come