Functional Characteristics of COPD Patients Admitted for Acute Pulmonary Embolism

C l i n M e d International Library Citation: Rodríguez DA, Orozco-Levi M, Miranda F, Mayoral A, Clements JA, et al., (2014) Functional Characteristics of COPD Patients Admitted for Acute Pulmonary Embolism. Int J Respir Pulm Med 1:003 Received: August 21, 2014: Accepted: September 08, 2014: Published: September 10, 2014 Copyright: © 2014 Rodríguez DA. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Rodríguez, Int J Respir Pulm Med 2014, 1:1 ISSN: 2378-3516


Introduction
Pulmonary embolism (PE) is a common cause of mortality, with an overall incidence rate of 69 cases per 100.000inhabitants [1].The clinical presentation and severity of PE are influenced by certain risk factors previously described in literature [2].During the last decade, there has been increased evidence that chronic obstructive pulmonary disease (COPD) is a risk factor for Venous Thrombo Embolism (VTE) [3,4] .Moreover, COPD patients present more frequently with PE than with deep venous thrombosis (DVT) [6,7], which has been shown to contribute to the poorer prognoses in these patients [5][6][7].Furthermore, similarities between the clinical manifestations of acute exacerbation of COPD (AECOPD) and PE have been shown to generate diagnostic challenges [8].Currently, there is very little information available with regards to the functional respiratory characteristics of COPD patients admitted for PE and/or the clinical variables capable of predicting the clinical course of these patients [7].The percentage of COPD patients admitted for PE does not exceed 10%, possibly due to the aforementioned diagnostic difficulties in this group [9,10].According to recent studies though, despite the low percentage of COPD cases admitted for PE, these patients still have a higher risk of mortality [7].Therefore, more information with regards to essential pulmonary functional characteristics and parameters of these patients would help to better diagnose and therefore rapidly identify therapeutic modalities for improved COPD patient survival.

Study design and measurements
We conducted a retrospective cohort study of 395 patients admitted for acute PE between January 2007 and December 2011 in a tertiary hospital setting.Patients were classified using the Wells criteria as having low, moderate or high risk for PE, as described previously [11].PEs were documented by either a positive helical computed tomography scan, a high-probability and/or an intermediateprobability ventilation-perfusion lung scan, a positive pulmonary angiography, or the visualization of a thrombus positioned in the right ventricle or right atrium on echocardiography [12].Deep vein thrombosis (DVT) was diagnosed following acute symptoms of DVT and confirmed by compression ultrasound or contrast venography of the lower extremities.Furthermore, complementary information was collected, such as: demographic data, symptoms at presentation, the type of diagnostic method used, risk factors for DVT and information pertaining to treatment and complications.In particular, major bleeding complications were defined as either bleeding requiring transfusion of two or more units of blood or a fatal bleed.Moreover, immobilized patients were categorized under two different categories: 1) non-surgical patients who had been immobilized for ≥4 days; 2) immobilized surgical patients', who had undergone a surgical procedure within last 2 months preceding PE development.
COPD was defined on the basis of smoking history and a post bronchodilator FEV1/FVC ratio less than 0.7 [13].We included COPD patients who were both clinically stable and whose diagnoses were made at least 2 months prior to admission for PE in order to avoid over diagnosis.

Data analysis
Results are expressed as the mean ± standard deviation (SD) for normally distributed variables.Categorical data are reported as numbers and percentages.Comparisons between subsets of COPD patients (survivor and non-survivor groups) were performed using an unpaired T-test for continuous variables and a chi-square test for categorical variables.Furthermore, as comparisons retained only Diffusion Lung Capacity (DLco) (expressed as percentage of predicted) [14], a Receiver Operating Characteristic (ROC) analysis was performed with mortality as the "gold standard" reference in order to determine the best cut-off point for DLco during the 3 month follow up period.. Likewise, the area under the curve (AUC) was calculated for ROC curve non-parametrically [15,16].The predictive values [17] were also calculated, both positive predictive value (PPV) and negative predictive value (NPV), in order to evaluate the best positive and negative results of the procedure.Afterwards, we explored the diagnostic capacity for prediction of DLco in the interval of 45% and 65% of the predicted value.Respective cut-off points were then selected that included the best sensitivity and specificity [18].We also evaluated the means and the 95% confidence intervals (95% CI) for sensitivity, specificity, PPV and NPV.Calculations were done with SPSS/PC (version 18.0, SPSS Inc., Chicago, IL, USA).A p-value of < 0.05 was considered significant.

Results
From January 2007 and December 2011, a total of 395 consecutive adult patients with objectively confirmed acute PE were included in the study.Of these, 33 (8.3%) were diagnosed with COPD at least 2 months prior to admission.
The total number of deaths after 3 months of follow-up was 65 (17%) in non-COPD patients and 9 (27%) in COPD patients with (p = 0.03).PE was the major cause of death for COPD patients (5 out of 9 deaths, 56%), while AECOPD (n=2) and lung cancer (n=2) represented 44% of COPD patient mortality.
Table 1 demonstrates and compares the main characteristics of our COPD patient groups (survivor and non-survivor groups) throughout the follow-up period. .On average, the overall sample of patients showed a severe airflow obstruction with 30% of these patients being current smokers.Up to 15% of COPD patients had frequent exacerbations.The principal risk factors that correlated with the development of PE were obesity, immobilization and active cancer, with a total of 10 cancer patients assessed in the study (6 lung, 2 bladder, 1 stomach and 1 pancreas).
Tachycardia and dyspnea were the most frequent presenting clinical symptoms.Comparisons between the individual subset groups (survivor and non-survivor) showed the DLco (%predicted) to be the only statically significan differing variable among these two group subsets.
Figure 1 includes a ROC curve with the best DLco (% predicted) cut-off points for the evaluation y during the 3-month follow-up period, including respective AUC calculations (AUC= 0.88; p = 0.001).
Table 2 demonstrates the values for sensitivity, PPV specificity and NPV for predicting mortality and indicates that a DLco equal to 60% is the threshold with the greatest capacity for predicting mortality (sensitivity: 0.88; PPV: 0.61; specificity: 0.80; NPV: 0.95).
Regarding initial PE treatment, one patient received thrombolytic therapy, while one other patient received an inferior vena cava filter.Initially, all patients were treated with low molecular weight heparin (LMWH) for 5 days.Afterwards, COPD patients continued treatment with VKA throughout the next 3 months.

Discussion
To our knowledge this is the first study reporting clinical and respiratory functional characteristics of COPD patients admitted for PE.The main results demonstrate that while clinically stable,

ISSN: 2378-3516
the pulmonary function parameters of these patients preceding PE development show: 1) severe airflow limitation and 2) a significant reduction of diffusion lung capacity for carbon monoxide in the non-survivor COPD patient who died during the 3-months followup compared to the COPD survivor group.Another important observation was that only 15% of COPD patients admitted for PE had frequent exacerbations.
Even though COPD patients with pe have been identified as a group of high risk mortality [7,8] , high risk bleeding and high risk vte recurrence [19] when compared to non-COPD patients developing pe, the clinical and pulmonary functional characteristics and parameters have still not been well explored in this population [7].
In our patients, low fev 1 , frequent exacerbations and/or increased comorbidities [13] did not represent high-risk factors.This study did identify; however, low dlco as being a a potential mortality risk factor.Low dlco has been generally described in patients with emphysema [13].A decrease in size of the gas exchange area of the lungs and ventilation-perfusion mismatching leads to a reduction in dl co in this subtype of COPD patients [20,21].Additionally, there are known associations between reduced dl co and clinical conditions such as acute pe [22,23].This underlying mechanism could furthermore involve a reduced blood volume in the pulmonary capillaries [24].Chronic pulmonary embolism, primary pulmonary hypertension (pph) and other pulmonary vascular diseases can also result in a decline in dl co [25].For these reasons, an objective reduction of dlco prior to pe admission may explain the results of this current study.
We consider various factors of our study to provide relevant implications not only for future research but as well as clinical management and stratification of COPD patients with pe.Firstly, the initial clinical evaluations of COPD patients, including the calculations for pe risk stratification, are usually based on classical scales [11,26].However, these scales do not take into consideration the severity of disease (i.e.Airflow limitation, dyspnea, etc.) Of COPD patients prior to pe development.This latter aspect was specifically researched throughout this study.Secondly, this study demonstrates that an adequate risk analysis could be beneficial for the improvement of individualized strategies on prevention, treatment and even follow-up following pe in this particular patient population [27].
The present study does though carry a series of limitations, among which is the size of the patient sample, its gender bias, as well as its retrospective nature.It should also be mentioned that lack of information concerning the degree of emphysema or presence of pulmonary hypertension could possibly have impacted the result interpretation of this study. .However, these limitations were also offset by two important strengths: 1) patients included in this study represented a homogenous group with a confirmed diagnosis COPD, which avoided possible over diagnosis; and 2) all of the pulmonary function studies were performed in the same laboratory, using a common systematic methodology.
In conclusion, the present findings showed that COPD patients admitted for pe have an elevated mortality when compared to non-COPD patients.Moreover, this study demonstrated for the first time, that COPD mortality from pe was associated with a manifested reduction in dl co prior to admission when compared to COPD survivors post-pe.
The present study constitutes a first attempt to increase our understanding of the complexity of pe pathogenesis in COPD patients.Future multicentric investigations though are warranted in order to confirm and expand on this study`s findings.

Table 2 :
Validity of the variables (means and 95% CI) for various DLco cut-points for predicting mortality.The value in bold indicates the selected cut-point.nPV= negative predictive value; PPV: positive predictive value.DLco: carbon monoxide diffusing capacity.

figure 1 :
figure 1: Receiver operating characteristic (ROC) curve (continuous line) with 95% CIs (dashed line) for the various DLco (as % predicted) cutoff points for death from any cause is shown.The best value (in bold) for mortality prediction from any cause was 60%.AUC: area under the curve.

Table 1 :
Clinical and functional characteristics of COPD patients admitted for acute pulmonary embolism (PE).