To the Editor: First termed by Adolf Kussmaul in an 1873 manuscript (1), the physical finding of pulsus paradoxus (PP) has been described in numerous clinical situations, including constrictive pericarditis, cardiac tamponade, acute pulmonary hypertension, severe asthma, tension pneumothorax, and exacerbations of chronic obstructive pulmonary disease (2). Originally described as the disappearance of the palpated pulse during inspiration in the setting of pericardial constriction, PP has more recently been defined as a drop in systolic blood pressure (SBP) of >10 mm Hg with inspiration. This classic physical finding is discussed at all levels of medical training and is frequently used in clinical medicine.

Obesity, defined as a body mass index (BMI) >30 kg/m², is a condition affecting over 30% of the U.S. population and is a global epidemic (3). In obese patients, the compressive effects of increased abdominal girth on the chest wall and diaphragm might increase the work of breathing. We hypothesized that this exaggerated respiratory effort might lead to PP in otherwise healthy obese patients. Accordingly, we performed a prospective study to investigate the relationship between obesity and PP in patients undergoing elective cardiac catheterization.

Adult patients presenting for elective cardiac catheterization were prospectively studied for the presence of PP. These were patients with no known or suspected pericardial or pulmonary diseases by complete history and chart review. Inclusion criteria were: adult patients undergoing elective cardiac catheterization who consented to participate in the study. Exclusion criteria were: 1) any known cause of PP, to include history of chronic obstructive pulmonary disease, active or chronic pericardial diseases, active asthma exacerbation, or use of any bronchodilating medications; 2) urgent need for catheterization (e.g., ST-segment elevation myocardial infarction or hemodynamic instability); 3) right ventricular infarction; 4) recent pulmonary embolism; 5) pregnancy; or 6) decompensated heart failure.

On the day of cardiac catheterization, a physical examination was performed to exclude the presence of pulmonary or pericardial disease. Height, weight, and body circumferences at the umbilical and xiphoid levels were measured. A limited two-dimensional transthoracic echocardiogram was performed to exclude occult pericardial effusion and to assess for structural changes consistent with constrictive pericardial disease. No patient was noted to have occult pericardial effusion or significant structural enlargement of the heart chambers. Pulsus paradoxus was first measured non-invasively with sphygmomanometry and then invasively assessed within 1 h at the time of cardiac catheterization. By convention, a PP value of >10 mm Hg was considered abnormal.

With a manual sphygmomanometer, repeated measurements of PP were taken in the supine position during normal resting tidal respirations until a single consistent value was determined. Invasive arterial pressure recordings were obtained via a standard coronary arteriographic catheter placed in the ascending aorta, recording for 5 to 10 respiratory cycles at 6.25 mm/s paper speed during normal resting tidal respirations. The change in SBP from inspiration to expiration was manually measured by an investigator unaware of the patients’ BMI. An average of the SBP change over a minimum of five respiratory cycles was recorded as the PP value for each patient. All study patients were in a regular rhythm.

The prespecified primary analysis was the prevalence of an abnormal PP value, defined as >10 mm Hg in the obese (BMI >30 kg/m²) versus the non-obese (BMI ≤30 kg/m²) study patients. Assuming a prevalence of obesity of 50% and a 5% prevalence of PP in non-obese subjects, the study of 100 patients was powered to detect a 20% difference in the prevalence of an abnormal PP between the obese and non-obese groups. The t test for independent groups was used for continuous variables, the chi-square test was used for comparison of categorical variables, and the Fisher exact test was used to evaluate the differences in gender between the two groups. Bivariate correlations between invasive PP values and abdominal girth (xiphoid and umbilical), body surface area, and BMI were performed with the Pearson correlation coefficient. Multivariate analysis was performed by multivariate linear regression.

(Table 1) demonstrates the characteristics and results for the obese and non-obese patients. The prevalence of an abnormal PP value (>10 mm Hg) was significantly higher in obese compared with non-obese patients (Figure 1A). Measured invasively, 46% of obese patients had PP versus 20% of non-obese patients (p = 0.012, odds ratio [OR] 3.3, 95% confidence interval [CI] 1.4 to 8.1). Measured non-invasively, 20% of the obese patients had PP versus 1.6% of the non-obese patients (p = 0.013, OR 15.5, 95% CI 1.8 to 132.0). With an alternate invasive criterion for PP defined as a 9% inspiratory drop in SBP (4), obesity was significantly associated with PP (OR 5.9, 95% CI 2.0 to 17.7, p = 0.002). The mean PP value was higher in the obese versus non-obese subjects with either invasive (10.1 ± 2.1 vs. 7.8 ± 1.5, p = 0.002) or non-invasive methods (7.4 ± 3.6 vs. 5.1 ± 2.1, p = 0.001) (Figure 1B). The PP values were significantly correlated with BMI (r = 0.28, p = 0.005), abdominal girth at the umbilical level (r = 0.33, p = 0.001), and xiphoid level (r = 0.27, p = 0.006). In multivariate analysis, umbilical girth was the strongest predictor of an elevated PP value.

Our study finds that the classic teaching of a PP value >10 mm Hg as an indicator of a pericardial or pulmonary disease process should also include obesity as a commonly associated finding. Other than case examples (5), this is the first study that systematically demonstrates a relationship between obesity and the increased prevalence of PP. This has clinical implications when dyspneic obese patients are evaluated for pericardial or pulmonary diseases. Failure to recognize this association might lead to misdiagnosis, unnecessary procedures, and additional hospital admissions (e.g., the obese asthmatic patient with dyspnea and a PP value >14).

Our study calls into question the accepted value of 10 mm Hg as the cutoff between an abnormal and normal PP value in obese patients. This analysis suggests that in using non-invasive sphygmomanometric methods, a value up to 16 mm Hg might be seen in otherwise healthy obese patients. By invasive measurements, the PP values exceeded 10 mm Hg in almost 50% of the obese patients and 20% of the non-obese patients. The question of PP values in disease states has been invasively studied in patients with known pericardial tamponade. In one of the largest series of patients with known pericardial tamponade, Reddy and Curtiss (4) proposed different criteria for PP consisting of an absolute inspiratory change of 12 mm Hg or >9% inspiratory drop in SBP. Applying these invasive criteria to our population without pericardial tamponade, we also found them to demonstrate an association between PP and obesity.

In summary, this study has shown that, with non-invasive or invasive assessment, an “abnormal” PP value is seen in a substantial proportion of individuals with a BMI exceeding 30 kg/m². Increasing abdominal girth also correlates with elevation in PP values. This might also impact the use of Doppler mitral inflow and other echocardiographic parameters of pericardial disease assessment in obesity. Higher thresholds are needed to avoid false positive diagnoses in obese patients with suspected pulmonary or pericardial diseases. Further study involving diseased and healthy obese populations is needed to fully define the optimal diagnostic values for an abnormal PP.