PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Am J Cardiol. Author manuscript; available in PMC 2010 July 15.
Published in final edited form as:
PMCID: PMC2787196
NIHMSID: NIHMS132365

Value of Orthopnea, Paroxysmal Nocturnal Dyspnea, and Medications in Prospective Population Studies of Incident Heart Failure

O. James Ekundayo, MD, DrPH,a Virginia J. Howard, PhD,a Monika M. Safford, MD,a Leslie A. McClure, PhD,a Donna Arnett, PhD,a Richard M. Allman, MD,a,b George Howard, DrPH,a and Ali Ahmed, MD, MPHa,b,*

Abstract

Prospective population studies of incident heart failure (HF) are often limited by difficulties in assembling a HF-free cohort. We used public-use copies of the Cardiovascular Health Study (CHS) datasets to determine sensitivity, specificity, and positive and negative predictive values (PPV and NPV) of orthopnea and paroxysmal nocturnal dyspnea (PND), with and without the use of medications used in CHS HF criteria (diuretics plus digoxin or vasodilators) in the diagnosis of prevalent HF and in the assembly of a relatively HF-free population. Of the 5771 community-dwelling older adults ≥65 years, 803 had orthopnea, 660 had PND, 1075 had either symptom, 388 had both symptoms, 547 were using HF medications, and 4315 had neither symptom nor were using HF medications. Definite HF was centrally adjudicated in 272 participants. The sensitivity, specificity, PPV and NPV (95% confidence intervals) for either orthopnea or PND were 52% (46%–58%), 83% (82%–84%), 13% (11%–15%) and 97% (97%–98%) respectively, and those for either orthopnea or PND or use of HF medications were 77% (72%–82%), 77% (76%–79%), 14% (13%–16%) and 99% (98%–99%) respectively. In conclusion, only <20% of those with either orthopnea or PND had definite HF, which limits their usefulness in the diagnosis of prevalent HF in the community. However, nearly 99% (NPV) of those with neither symptom nor using HF medications also did not have HF, which may be useful as a simple and inexpensive tool in assembling a relatively HF-free cohort for prospective population studies of incident HF.

Keywords: Orthopnea, paroxysmal nocturnal dyspnea, medications, heart failure, diagnosis, population studies

Heart failure (HF) is often difficult to diagnose in population studies.1, 2 This complicates population studies of incident HF as participants should ideally be free of prevalent HF at baseline. In clinical settings, orthopnea and paroxysmal nocturnal dyspnea (PND) are often considered specific symptoms of HF.1, 3 HF medications have also been used to diagnose HF in population settings. Although data on these symptoms and HF medications may be collected in population studies, their role in the diagnosis of prevalent HF in this setting is relatively unknown. The objective of this study was to determine the usefulness of orthopnea and PND, with and without HF medications, in assembling a cohort relatively free of prevalent HF using public-use copies of the Cardiovascular Health Study (CHS) datasets obtained from the National Heart, Lung, and Blood Institute (NHLBI).

Methods

CHS is an ongoing, prospective epidemiologic study of cardiovascular risk factors funded by the NHLBI. A total of 5888 community-dwelling older adults ≥65 years were recruited from a random sample of Medicare-eligible residents from the four study sites: Sacramento County, California; Washington County, Maryland; Forsyth County, North Carolina; and Allegheny County, Pennsylvania.4 An original cohort of 5201 participants was recruited between 1989 and 1990 and a second cohort of 687 African-Americans was recruited between 1992 and 1993. We used public-use copies of the CHS datasets obtained from the NHLBI, which contained data from 5795 participants, as some participants did not consent to be included in the de-identified public-use copy of the data. Of the 5795 participants, 5771 had data on baseline orthopnea or PND and were used in the current analysis.

Data on orthopnea and PND were collected during baseline home interviews using a CHS questionnaire. For orthopnea, participants were asked “Have you ever had to sleep on two or more pillows to help you breathe” and for PND, the question was “Have you ever been awakened at night by trouble breathing.” Data on use of prescription medications during the preceding two weeks were collected directly from prescription bottles, and data on the use of nonprescription medications such as aspirin, sleeping pills and antihistamines were collected using questionnaires.4 Medications used in the criteria for the diagnosis of HF included the use of any diuretics, and either digoxin or a vasodilator including an angiotensin-converting enzyme (ACE) inhibitor.

Of the 5771 participants, 272 (4.7%) had definite baseline HF as centrally adjudicated by a committee and constituted “gold standard” cases for these analyses. The CHS criteria for HF assessment are well-described in the literature and have been shown to be more stringent than the Framingham criteria for HF.410 Briefly, a panel of experts adjudicated prevalent HF by reviewing all relevant data including history, physical examination, chest x-ray and medications from medical records related to hospitalizations or outpatient visits. The process began with self-reports of physician-diagnosed HF, which was then adjudicated by a central CHS committee of physicians based on the receipt of medical therapy for HF (defined as current use of a diuretic and either digitalis or a vasodilator, namely, ACE inhibitor, nitroglycerin, or hydralazine).5 For those with self-reported, physician-diagnosed HF who did not meet the medication criteria, a definite diagnosis of HF was made after medical record review for symptoms (shortness of breath, fatigue, orthopnea, PND), signs (edema, pulmonary râles, gallop rhythm, displaced left ventricular apical impulse), and tests (chest radiography).5 When self-reported, physician-diagnosed HF could not be confirmed using the above approaches, CHS field centers sought confirmation from treating physicians by questionnaires or from hospitals by discharge summaries which were then reviewed for symptoms, signs, and chest x-ray findings suggestive of HF. However, some cases of prevalent HF at baseline were retrospectively confirmed through surveillance during follow-up.5 All information was reviewed centrally at the CHS coordinating center for consistency of classification among the four centers.

Considering the clinical relevance of the symptoms of orthopnea and PND, we at first sought to understand the characteristics of participants who had either orthopnea or PND at baseline compared to those who had neither. As such, we categorized participants into two groups based on the presence or absence of either of these two symptoms and compared them using Pearson’s chi-square and Student’s t-tests as appropriate. We estimated sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) along with 95% confidence intervals (CI) of baseline presence of orthopnea or PND for committee-adjudicated definite HF.11, 12 We then repeated our analysis adding HF medications into the criteria (orthopnea or PND or the use of HF medications). Data were analyzed using SPSS for Windows release 15 (Chicago, IL, 2008).

Results

Participants (n=5771) had a mean age of 73 (±6.0) years, 57% were women and 16% were African Americans. At baseline, orthopnea and PND was reported by 803 (13.9%) and 660 (11.4%) participants, respectively, 1075 (18.6%) of participants reported either symptom, and 547 (9.5%) were using HF medications. Patients with either orthopnea or PND were more likely to be women, nonwhites, unmarried, living alone, have higher mean body mass index, higher comorbidity burden and report fair to poor health (Table 1). Only 388 (6.7%) participants had both symptoms. A total of 1456 (25.2%) participants had either orthopnea or PND, or received HF medications, and 4315 had neither symptom nor were using HF medications.

Table 1
Baseline characteristics by either orthopnea or paroxysmal nocturnal dyspnea (PND)

The sensitivity, specificity, PPV and NPV for either orthopnea or PND at baseline were respectively 51.8%, 83.0%, 13.1% and 97.2% (Table 2). The values for individual symptoms of orthopnea and PND, and combination thereof, are displayed in Table 2. The sensitivity, specificity, PPV and NPV for either orthopnea or PND or the use of HF medications at baseline were respectively 77.2%, 77.3%, 14.4% and 98.6% (Table 2 and Figure 1). The sensitivity, specificity, PPV and NPV for either orthopnea or PND or the use of HF medications at baseline for subgroups of participants are displayed in Table 3.

Figure 1
Two by two table demonstrating sensitivity, specificity, and predictive values of symptoms of orthopnea or paroxysmal nocturnal dyspnea and HF medications in the Cardiovascular Health Study. (CI = confidence interval; PPV = positive predictive value or ...
Table 2
Sensitivity, specificity, positive predictive values, and negative predictive values, along with 95% confidence interval (CI) of orthopnea and paroxysmal nocturnal dyspnea, with and without heart failure (HF) medications for the diagnosis of HF in the ...
Table 3
Sensitivity, specificity, positive predictive values, and negative predictive values, along with 95% confidence intervals (CI) of either orthopnea or paroxysmal nocturnal dyspnea or heart failure medications for the diagnosis of heart failure in subgroups ...

Discussion

The findings of our study demonstrate that the prevalence of symptoms of orthopnea or PND was relatively common among community-dwelling older adults. Over 97% of CHS participants who did not report either orthopnea or PND (NPV), also did not have HF. However, when CHS criteria for HF medications was added, nearly 99% of those who reported neither orthopnea nor PND or used HF medications, also did not have HF, suggesting that these criteria can be used to identify a cohort relatively free of prevalent HF. These findings are important as they provide simple and clear evidence for the use of an inexpensive tool to exclude prevalent HF patients in community cohorts, an important requirement for prospective population studies of incident HF.

Orthopnea and PND are considered major Framingham criteria for the diagnosis of HF.13 The presence of two major Framingham criteria confirms a diagnosis of HF, and as such, CHS participants with both orthopnea and PND met the Framingham criteria for the diagnosis of HF. However, only 20% (PPV) of CHS participants who reported both orthopnea and PND had HF. The low prevalence of HF among CHS participants may in part explain the low PPV of these two symptoms, as PPV is directly related to prevalence. However, the 4.7% prevalence of HF in the CHS is likely to be one of the highest in a community setting as the prevalence of HF increases with age and the CHS participants were ≥65 years. Another possible explanation for the low PPV of these two symptoms is that they may also be manifestations of other conditions such as chronic obstructive pulmonary disease or sleep apnea, the prevalence of which increases with age. Thus, the low PPV of these two symptoms makes them unsuitable for the diagnosis of HF in the community setting.

This is in contrast to the usefulness of these symptoms in the clinical setting where follow-up questions may clarify their presence or absence, which may also be assessed in the context of other symptoms, signs, and other medical conditions, such as coronary artery disease or hypertension, which are underlying etiologic factor for most HF. Furthermore, the prevalence of HF in the clinical setting is expected to be higher, which is also likely to improve the PPV of these symptoms. Among patients presenting with dyspnea in the emergency department in whom the prevalence of HF is estimated to be as high as 50%, the PPV of orthopnea and PND is estimated to be 68% and 72% respectively.3, 14 The sensitivity and specificity of orthopnea and PND observed in our study among community-dwelling older adults were similar to those of patients with acute dyspnea.3 This suggests that these two symptoms alone may not be very reliable in the diagnosis of HF even in the clinical setting, and that coordination with other clinical information is necessary.

The absence of either orthopnea or PND, on the other hand, may be useful in identifying those without HF. Over 97% (NPV) of CHS participants without either symptom did not have HF. Yet, it may not be suitable to assemble a cohort free of HF, as the 3% of the participants with HF (false negative) would include nearly half of all prevalent HF patients. However, when HF medications were added into the criteria, nearly 99% of those with neither symptom nor receiving HF medications were free of HF. Importantly, the addition of HF medication increased the NPV from 97.2% to 98.6%, with 95% confidence intervals (98.2% to 98.9%) that did not include the NPV for orthopnea or PND (97.2%), thus highlighting the statistical significance of the improvement in NPV with the addition of HF medications. Although the remaining 1% of participants with HF includes about one fifth of all HF patients, they are asymptomatic and thus likely to be stable and may have better prognoses. Therefore, by collecting data on orthopnea, PND, and HF medications used in the CHS criteria for the diagnosis of HF, one can assemble a cohort that may include about 1% of asymptomatic stable HF patients, which may be suitable for long-term follow up of incident HF and HF hospitalization. The NPV is inversely related to prevalence. However, the effect of the prevalence of HF on NPV is likely to be minimal as the prevalence of HF in the community can only vary within a narrow range (between 1% and 8%).15 The 4.7% prevalence of HF in CHS would be considered high as the prevalence of HF increases with age and would likely be lower in cohorts younger than 65 years, thus yielding higher NPV.

The relevance or lack thereof of the sensitivity and specificity of a screening tool to assemble a HF-free cohort in the community setting deserves further discussion. In the clinical setting, sensitivity and specificity of diagnostic tests are determined in studies using gold standard diagnostic tests and those probabilities are used on individual patients, and are not meant to assemble cohorts. A diagnostic test should have high sensitivity (ability to detect a true positive) if the cost of missing cases (false negative) is too high to be acceptable. For example, if phenylketonuria is not detected early after birth, it may cause irreversible brain damage.11, 16 On the other hand, a test should have high specificity if the cost of wrong diagnoses (false positive) is too high to be acceptable. For example, a wrong diagnosis of prostate cancer may lead to anxiety, insurance consequences and potentially invasive diagnostic and therapeutic interventions. HF is a clinical syndrome, which unlike phenylketonuria or prostate cancer, is unlikely to be asymptomatic on initial presentation. Therefore, a high false negative rate (low sensitivity) may delay the diagnosis of HF and initiation of HF therapy, but is not likely to be life-threatening in those without the typical HF symptoms of orthopnea and/or PND. On the other hand, a high false positive rate (low specificity) may lead to false diagnosis of HF which in turn may lead to prescription of HF medications or even hospitalization. Therefore, a high specificity may be more desirable for HF screening in the clinical setting.

The modest 77% specificity for orthopnea or PND or HF medication use in the community in our study is comparable to the 76% specificity of brain natriuretic peptide (BNP) using a cutoff of 100 pg/ml observed in patients with dyspnea in the emergency department.17 In that study, the highest BNP cutoff of 150 pg/ml yielded a specificity of 83%. Of note, when we used either orthopnea alone, or PND alone, or the presence of both orthopnea and PND, we obtained much higher specificities (88%, 90% and 94% respectively). However in the clinical setting, a diagnosis of HF is complex and requires assessment of a set of clinical and laboratory findings, which may not be feasible in the community setting. Although there is no data on the usefulness of BNP in the diagnosis of HF in the community, BNP was not useful for community screening for left ventricular hypertrophy and systolic dysfunction in the Framingham Heart Study.18 Therefore, the criteria used in our study provide an inexpensive yet efficient tool for HF screening in the community for population studies of incident HF.

One of the limitations of our study is that the participants were older adults. Because the prevalence of HF may be lower in a younger population, the findings of this study may not be generalizable to a younger population. However, over 80% of HF patients are ≥65 years, and data from our subgroup analyses suggest that the NPV was similar regardless of age. Also, unlike PPV, which is reduced with lower prevalence, the NPV increases with the decrease in prevalence, therefore the findings of this study may be more useful in assembling a younger HF-free population.19 Another limitation was that we had no data on angiotensin receptor blockers or beta blockers currently recommended for use in HF. None of the CHS participants were receiving angiotensin receptor blockers. Data on beta blockers were also collected but were not included in the CHS HF medication list. Beta blockers were considered harmful and contraindicated in HF during the period when CHS participants were being enrolled. The results of the major randomized trials of beta blockers in HF that are currently recommended for use in HF (carvedilol, metoprolol succinate extended release and bisoprolol) were published after CHS enrollment. Future studies need to examine if the inclusion of these drugs would improve the CHS HF criteria. In conclusion, when used in conjunction with HF medications, orthopnea and PND can identify nearly 99% of those without HF in the community, and thus this simple, efficient and inexpensive tool can be useful in assembling relatively HF-free cohorts for prospective studies of incident HF.

Acknowledgments

Funding/Support: Dr. Ahmed is supported by the National Institutes of Health through a grant from the National Heart, Lung, and Blood Institute (R01-HL085561) and a generous gift from Ms. Jean B. Morris of Birmingham, Alabama. Dr. Safford is supported by NIH R01 HL80477-01A1 (NHLBI).

“The Cardiovascular Health Study (CHS) was conducted and supported by the NHLBI in collaboration with the CHS Investigators. This manuscript was prepared using a limited access dataset obtained by the NHLBI and does not necessarily reflect the opinions or views of the CHS Study or the NHLBI.”

Footnotes

Conflict of Interest Disclosures: None

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

1. Ahmed A. DEFEAT heart failure: clinical manifestations, diagnostic assessment, and etiology of geriatric heart failure. Heart Fail Clin. 2007;3:389–402. [PubMed]
2. Luepker R, Benjamin E, Mensah G, O’Connell J, Pfeffer M, Psaty BM. Minutes of Meeting. Paper presented at: Special Emphasis Panel on the Standardized Assessment of Heart Failure in Population Studies; Bethesda, Maryland. 1997. Available at: http://www.nhlbi.nih.gov/meetings/workshops/hfpopmin.txt.
3. Wang CS, FitzGerald JM, Schulzer M, Mak E, Ayas NT. Does this dyspneic patient in the emergency department have congestive heart failure? JAMA. 2005;294:1944–1956. [PubMed]
4. Fried LP, Borhani NO, Enright P, Furberg CD, Gardin JM, Kronmal RA, Kuller LH, Manolio TA, Mittelmark MB, Newman A, et al. The Cardiovascular Health Study: design and rationale. Ann Epidemiol. 1991;1:263–276. [PubMed]
5. Psaty BM, Kuller LH, Bild D, Burke GL, Kittner SJ, Mittelmark M, Price TR, Rautaharju PM, Robbins J. Methods of assessing prevalent cardiovascular disease in the Cardiovascular Health Study. Ann Epidemiol. 1995;5:270–277. [PubMed]
6. Gottdiener JS, Arnold AM, Aurigemma GP, Polak JF, Tracy RP, Kitzman DW, Gardin JM, Rutledge JE, Boineau RC. Predictors of congestive heart failure in the elderly: the Cardiovascular Health Study. J Am Coll Cardiol. 2000;35:1628–1637. [PubMed]
7. Kitzman DW, Gardin JM, Gottdiener JS, Arnold A, Boineau R, Aurigemma G, Marino EK, Lyles M, Cushman M, Enright PL. Importance of heart failure with preserved systolic function in patients > or = 65 years of age. CHS Research Group. Cardiovascular Health Study. Am J Cardiol. 2001;87:413–419. [PubMed]
8. Gottdiener JS, McClelland RL, Marshall R, Shemanski L, Furberg CD, Kitzman DW, Cushman M, Polak J, Gardin JM, Gersh BJ, Aurigemma GP, Manolio TA. Outcome of congestive heart failure in elderly persons: influence of left ventricular systolic function. The Cardiovascular Health Study. Ann Intern Med. 2002;137:631–639. [PubMed]
9. Schellenbaum GD, Rea TD, Heckbert SR, Smith NL, Lumley T, Roger VL, Kitzman DW, Taylor HA, Levy D, Psaty BM. Survival associated with two sets of diagnostic criteria for congestive heart failure. Am J Epidemiol. 2004;160:628–635. [PubMed]
10. Schellenbaum GD, Heckbert SR, Smith NL, Rea TD, Lumley T, Kitzman DW, Roger VL, Taylor HA, Psaty BM. Congestive heart failure incidence and prognosis: case identification using central adjudication versus hospital discharge diagnoses. Ann Epidemiol. 2006;16:115–122. [PubMed]
11. Ahmed A, Allman RM, Aronow WS, DeLong JF. Diagnosis of heart failure in older adults: predictive value of dyspnea at rest. Arch Gerontol Geriatr. 2004;38:297–307. [PubMed]
12. Hamm RM. Clinical Decision Making Calculators.: Department of Family Medicine, University of Oklahoma. Health Sciences Center; [Accessed on July 15, 2008]. 2008. available on the Internet at http://www.fammed.ouhsc.edu/robhamm/cdmcalc.htm#Decision.
13. Senni M, Tribouilloy CM, Rodeheffer RJ, Jacobsen SJ, Evans JM, Bailey KR, Redfield MM. Congestive heart failure in the community: a study of all incident cases in Olmsted County, Minnesota, in 1991. Circulation. 1998;98:2282–2289. [PubMed]
14. Mueller C, Scholer A, Laule-Kilian K, Martina B, Schindler C, Buser P, Pfisterer M, Perruchoud AP. Use of B-type natriuretic peptide in the evaluation and management of acute dyspnea. N Engl J Med. 2004;350:647–654. [PubMed]
15. Kupari M, Lindroos M, Iivanainen AM, Heikkila J, Tilvis R. Congestive heart failure in old age: prevalence, mechanisms and 4-year prognosis in the Helsinki Ageing Study. J Intern Med. 1997;241:387–394. [PubMed]
16. Hennekens CH, Buring JE. Epidemiology in Medicine. Boston/Toronto: Little, Brown and Co; 1987.
17. Maisel AS, Krishnaswamy P, Nowak RM, McCord J, Hollander JE, Duc P, Omland T, Storrow AB, Abraham WT, Wu AH, Clopton P, Steg PG, Westheim A, Knudsen CW, Perez A, Kazanegra R, Herrmann HC, McCullough PA. Rapid measurement of B-type natriuretic peptide in the emergency diagnosis of heart failure. N Engl J Med. 2002;347:161–167. [PubMed]
18. Vasan RS, Benjamin EJ, Larson MG, Leip EP, Wang TJ, Wilson PW, Levy D. Plasma natriuretic peptides for community screening for left ventricular hypertrophy and systolic dysfunction: the Framingham heart study. JAMA. 2002;288:1252–1259. [PubMed]
19. Akobeng AK. Understanding diagnostic tests 1: sensitivity, specificity and predictive values. Acta Paediatr. 2007;96:338–341. [PubMed]