Although P-wave axis is routinely reported on the printouts of most contemporary electrocardiograms, the prognostic significance of its abnormality has not been systematically evaluated. We examined the association between abnormal P-wave axis and cardiovascular and all-cause mortality in 7,501 participants (53% women, mean age 59 years) from the United States Third National Health and Nutrition Examination Survey. P-wave axis of 0° to 75° was considered normal. Participants were linked to the National Death Index to identify the underlying cause of death listed on the death certificates using the International Classification of Disease. During a median follow-up of 13.8 years, a total of 2,541 deaths occurred; of which 1,077 were due to a cardiovascular cause. Abnormal P-wave axis was associated with 55% increased risk of all-cause mortality (hazard ratio [HR] 1.55, 95% confidence interval [CI] 1.43 to 1.69, p <0.01) and 41% increased risk of cardiovascular mortality (HR 1.41, 95% CI 1.24 to 1.62, p <0.01). After adjustment for age, gender, race/ethnicity, diabetes, systolic blood pressure, body mass index, smoking status, total/high-density lipoprotein cholesterol ratio, previous cardiovascular disease, chronic obstructive pulmonary disease, bronchial asthma, heart rate, and use of antiarrhythmic or atrioventricular nodal blocking drugs, the risk of mortality remained significantly high (HR 1.24 95% CI 1.13 to 1.36, p <0.01 for all-cause mortality and HR 1.19 95% CI 1.03 to 1.38, p = 0.02 for cardiovascular mortality) and was consistent across several subgroups of the participants. In conclusion, abnormal P-wave axis is associated with an increased risk of death, findings that call for attention to this routinely reported finding on contemporary electrocardiographic machines.
P-wave axis is one of the routinely reported measurements on the printout of the 12-lead electrocardiogram (ECG) obtained by most current electrocardiographs. Generally, physicians pay little or no attention to it. This is possibly due to lack of data on the prognostic significance of abnormal P-wave axis. Apart from its classic use in screening for chronic obstructive pulmonary disease (COPD), there is almost no literature on the long-term outcomes associated with abnormal P-wave axis. Studies on the long-term outcomes associated with abnormal P-wave axis are warranted and could aid in screening and risk stratification in asymptomatic subjects with subclinical cardiovascular and/or pulmonary disease. We examined the association between abnormal P-wave axis and cardiovascular and all-cause mortality in the United States Third National Health and Nutrition Examination Survey (NHANES III).
Methods
The NHANES is a periodic survey of a representative sample of the civilian noninstitutionalized United States population. Its principal aim is to determine estimates of disease prevalence and health status of the United States population. The survey design, its components, and resulting data are available at the Web site of the US Centers for Disease Control and Prevention ( http://www.cdc.gov/nchs/nhanes/nh3data.htm ).
The present analysis includes 7,501 participants from NHANES III who had good-quality ECGs showing sinus rhythm and no major intraventricular conduction delay (including complete bundle branch blocks and/or QRS duration ≥120 ms) as well as available mortality data, medical history, medication use, and anthropometric measurements.
NHANES III baseline data were collected during an in-home interview and a subsequent visit to a mobile examination center from 1988 to 1994. The data collected during the in-home interview included demographics, medical history including smoking status, and use of medications. Using the height and weight measured during the visit to the mobile examination center, the body mass index was calculated as the weight in kilograms divided by the height in meters squared. Blood pressure was measured 3 times during the in-home interview and 3 additional times during the visit to the mobile examination center. All blood pressure measurements for each participant were averaged. Diabetes mellitus was defined as a fasting plasma glucose of ≥126 mg/dl, a nonfasting plasma glucose of ≥200 mg/dl, and/or a self-reported history of diabetes with concurrent use of antidiabetes medication. The total serum cholesterol and high-density lipoprotein cholesterol levels were measured enzymatically.
Standard 12-lead ECG was recorded on a Marquette MAC 12 system (Marquette Medical Systems, Milwaukee, Wisconsin) by trained technicians during the participant’s visit to a mobile examination center. Computerized automated analysis of the electrocardiographic data was performed, which included selective averaging to obtain the representative durations and amplitudes of electrocardiographic components and frontal axis. P-wave axis with value between 0° and 75° was considered normal.
The NHANES III participants have been followed up for mortality through December 31, 2006. The method of probabilistic matching was used to link the NHANES III participants with the National Death Index to identify vital status and, for those who died, the cause of death. Matching was based on 12 identifiers for each participant, including Social Security number, gender, and date of birth. The follow-up period for each study participant was calculated as the interval between their NHANES III examination and the date of death or December 31, 2006, whichever occurred first. The cause of death was determined using the underlying cause listed on the death certificates. The “International Classification of Disease” (ICD), ninth revision, was used for deaths occurring from 1988 to 1998 and ICD-10 for deaths occurring from 1999 to 2006. Cardiovascular mortality was defined by any of the ICD-9 codes 390 to 434 and 436 to 459 and ICD-10 codes I00 to I99.
Frequency distributions of all variables were first inspected to identify anomalies and outliers possibly caused by measurement artifacts. Continuous data are described by their mean and SD and categorical data as proportions (percentage). Differences in characteristics by P-wave axis status (abnormal vs normal) were assessed by chi-square (for categorical variables) and unpaired t (for continuous variables) tests.
Cox proportional hazards analysis was used to calculate the unadjusted and multivariate-adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) for all-cause and cardiovascular mortality, separately associated with abnormal P-wave axis in a series of incremental models as follows: first unadjusted (model 1); then adjusted for demographics (age, gender, and race/ethnicity) (model 2); and then additionally adjusted for diabetes, systolic blood pressure, body mass index, smoking status, total/high-density lipoprotein cholesterol ratio, previous cardiovascular disease, COPD, bronchial asthma, heart rate, and use of antiarrhythmic or atrioventricular nodal blocking drugs (model 3). The covariates in the models were selected because of their association with P-wave axis and being potential confounders. To examine the impact of common electrocardiographic intervals on the associations between abnormal P-wave axis and outcomes, we adjusted for P duration, PR interval, QT duration, and QRS duration in additional models.
In models similar to model 3, we also examined the association between abnormal P-wave axis and all-cause and cardiovascular mortality, separately, across subgroups of the study participants stratified by age, gender, race/ethnicity, COPD status, and history of cardiovascular disease.
All analyses were done using SAS 9.3 (SAS Institute Inc., Cary, North Carolina). Statistical significance was determined as a 2-sided p <0.05.
The National Center for Health Statistics of the Center for Disease Control and Prevention institutional review board approved the protocol for NHANES III. All participants gave written informed consent. The investigators are solely responsible for the design and conduct of this analysis as well as drafting and editing of the report and its final contents.
Results
A total of 7,501 participants (53% women, age 59.3 ± 13.3 years) were included in this study. Abnormal P-wave axis was detected in 1,748 participants. Table 1 lists the characteristics of the study population by the status of P-wave axis. As shown, there were racial/ethnic but no gender differences in the distribution of abnormal P-wave axis. Participants with abnormal P-wave axis were more likely to be older, smokers, and with history of COPD and bronchial asthma. Diabetes and use of antiarrhythmic and atrioventricular nodal blocking drugs were more common in participants with normal P-wave axis. Heart rate and high-density lipoprotein cholesterol levels were higher but body mass index was lower in participants with normal P-wave axis ( Table 1 ).
| Variable | Normal P axis (n = 5,753) | Abnormal P axis (n = 1,748) | p |
|---|---|---|---|
| Age (yrs) | 58.6 ± 13.1 | 61.6 ± 13.6 | <0.01 |
| Women | 3,044 (52.9) | 954 (54.6) | 0.22 |
| Race/ethnicity | <0.01 | ||
| White | 2,706 (47.4) | 966 (55.3) | |
| Black | 1,296 (22.5) | 455 (26.0) | |
| Mexican | 1,508 (26.2) | 253 (14.5) | |
| Others | 243 (4.2) | 74 (4.2) | |
| Systolic blood pressure (mm Hg) | 132.9 ± 19.6 | 131.8 ± 20.4 | 0.06 |
| Total cholesterol (mg/dl) | 219 ± 44 | 216 ± 45 | 0.01 |
| High-density lipoprotein cholesterol (mg/dl) | 50 ± 15 | 55 ± 18 | <0.01 |
| Total/high-density lipoprotein cholesterol ratio | 4.8 ± 1.7 | 4.3 ± 1.64 | <0.01 |
| Antiarrhythmic drug use | 961 (16.7) | 242 (13.8) | <0.01 |
| COPD | 401 (7.0) | 258 (14.8) | <0.01 |
| Bronchial asthma | 352 (6.1) | 188 (10.8) | <0.01 |
| Hear rate (beats/min) | 67.8 ± 11.0 | 70.32 ± 12.4 | <0.01 |
| Diabetes mellitus | 708 (12.3) | 152 (8.7) | <0.01 |
| Smoking status | <0.01 | ||
| Ever smoked | 3,025 (52.6) | 1,083 (61.9) | |
| Past | 1,886 (32.8) | 517 (29.6) | |
| Current | 1,139 (19.8) | 566 (32.4) | |
| Body mass index (kg/m 2 ) | 28.5 ± 5.4 | 24.9 ± 5.17 | <0.01 |
| P-wave axis (degrees) | 53.5 ± 16.9 | 76.8 ± 27.0 | <0.01 |
| Previous cardiovascular disease | 474 (8.2) | 136 (7.8) | 0.58 |
During a median follow-up of 13.8 years, a total of 2,541 deaths occurred (incidence 26.5 per 1,000 person-years); of which, 1,077 were due to a cardiovascular cause (incidence 11.3 per 1,000 person-years). In participants with abnormal P-wave axis, the rates of all-cause mortality (36.5 per 1,000 person-years) and cardiovascular mortality (14.5 per 1,000 person-years) were significantly higher compared with the rates in those with normal P-wave axis (23.8 per 1,000 person-years for all-cause mortality and 10.4 per 1,000 person-years for cardiovascular mortality, p <0.001). Figure 1 shows the event-free survival curves of participants with and without abnormal P-wave axis.
In Cox proportional hazard analysis, abnormal P-wave axis was associated with 55% increased risk of all-cause mortality (<0.001) and 41% increased risk of cardiovascular mortality (p <0.001). After adjustment for several potential confounders, the risk of mortality remained significantly high but was slightly attenuated as more covariates were entered in the models ( Table 2 ). Further adjustment for P duration, PR interval, QT duration, and QRS duration did not significantly attenuate the risk for all-cause mortality (HR 1.22, 95% CI 1.11 to 1.34, p <0.001) or cardiovascular mortality (HR 1.17, 95% CI 1.01 to 1.35, p = 0.039).
