Depression profile in patients with and without chronic heart failure

1. Introduction 

Chronic heart failure (CHF) is a common and disabling disease with poor prognosis (Muntwyler et al., 2002). Its typical symptom triad is dyspnea, limited physical capacity and peripheral edema, which in part explains the severely reduced quality of life of patients with chronic heart failure (Jünger et al., 2002, Konstam et al., 1996). Unfortunately, chronic heart failure is often complicated by comorbid depression: While 4 to 10% of the general population in the United States and Europe meet diagnostic criteria for depressive disorders (12-month prevalence rates) (Demyttenaere et al., 2004, Kessler et al., 2003), 11 to 25% of CHF outpatients and 35 to 70% of CHF inpatients are estimated to suffer from depressive disorders (O'Connor and Joynt, 2004, Rutledge et al., 2006). Depression is a strong predictor of short-term worsening of heart failure symptoms (Rumsfeld et al., 2003) and increased mortality (Jünger et al., 2005). Despite these high prevalence rates and the important significance of depression for the course of physical illness, depression often goes unrecognized and untreated in clinical practice (Davidson and Meltzer-Brody, 1999, Löwe et al., 2004c). The poor recognition rates of depression in CHF patients might be explained by the symptom overlap of the two disorders: CHF and depression are both characterized by fatigue, loss of energy, poor appetite, sleep disturbances, psychomotor retardation, and concentration deficits (Simon and von Korff, 2006). This symptom overlap may hamper the diagnosis and monitoring of depression symptoms during the CHF treatment process. Specific risk factors for depression in CHF patients, such as for example higher NYHA functional class (Rutledge et al., 2006, Holzapfel et al., 2007), younger age (Freedland et al., 2003) or limited quality of life (Gottlieb et al., 2004), can help to identify patients with depression. However, in order to improve the diagnosis of depression in CHF patients, an awareness of the specific symptom profile of depression as presented in CHF patients is necessary. This also applies to depression in other chronic diseases, for example in diabetes mellitus, asthma or chronic obstructive lung disease (Cavanaugh et al., 1983). A number of studies have compared depressive symptom patterns in patients with and without somatic comorbidity (Simon and von Korff, 2006), or in patients recruited in medical and psychiatric settings (Clark et al., 1998). So far the findings have remained ambiguous: While an earlier study concluded that there is no clinically relevant difference in the depression symptom profile in medical and non-medical populations (Coulehan et al., 1988), another study found that the features discriminating depressed patients with and without comorbid physical illness were cognitive symptoms such as hopelessness, suicidal ideation and feelings of guilt (Clark et al., 1998). In a recent study, Simon and von Korff (2006) found only modest differences in the severity of somatic depression symptoms in depressed patients with and without medical comorbidity, leading to the conclusion that the DSM-IV diagnostic criteria do not require modification for patients with medical comorbidity. Finally, Belkin et al. stated that a marked increase in somatic symptoms is an indicator for more severe depression (Belkin et al., 1992). There are no studies investigating the profile of depression specifically in CHF patients.

Given the heterogeneity of the findings cited above, and the lack of studies describing the specific profile of depression in patients with CHF as well as its clinical relevance for diagnosis and treatment, we sought to describe the specific depression profile of patients with CHF. To detect depressive symptom patterns unique to patients with CHF, we compared the severity of individual DSM-IV depressive symptoms in depressed patients with and without CHF.

2. Methods 

3. Results 

3.1. Description of patients 

The 677 patients from the CHF outpatient clinic ranged in age from 16 to 90 years (M=57.2, SD=13.8 years); 76.1% were male. All patients from this sample suffered from documented CHF. Forty-two patients (6.2%) met the diagnostic criteria for major depressive disorder, and 71 patients (10.5%) met the diagnostic criteria for other depressive disorders according to the PHQ-9 diagnostic algorithm.

Two hundred forty-eight patients from the psychosomatic outpatient clinic participated in this study. To compile a comparison group without CHF, 4 patients were excluded because a heart disease was documented in their medical records. None of the remaining 244 patients had evidence of CHF in their medical history and record. The age range was 16 to 79 years (M=39.1, SD=15.1); 28.8% of the patients were male. Ninety-five patients (38.9%) met the diagnostic criteria for major depressive disorder, and 42 patients (17.2%) met the diagnostic criteria for other depressive disorders according to the PHQ-9 diagnostic algorithm.

Table 1 summarizes the sociodemographic characteristics and medical comorbidities for each of the 4 patient subsamples. As expected, the subsamples of CHF and non-CHF patients differed with respect to age (major depressive disorder: t(133)=−8.5, p≤.001, other depressive disorder: t(107)=−5.9, p≤.001) and gender (major depressive disorder: χ12=35.9, p≤.001, other depressive disorder: χ12=11.1, p=.001), with the CHF patients being significantly older and more frequently male compared to the patients without CHF.

Table 1. Sociodemographic data and medical comorbidities

		Major Depression		Other Depression
		CHF n=42	Non-CHF n=95	CHF n=71	Non-CHF n=42
	Mean age, years, M (SD)	56.5 (14.2)	35.2 (13.2)	60.1 (14.4)	42.2 (16.3)
	Gender, male, %	73.8	20.0	63.4	31.0
	Chronic heart failure, %	100	–	100	–
	°Dilative cardiomyopathy	56.1	–	58.0	–
	°Coronary heart disease	39.0	–	29.0	–
	°Other	4.9	–	13	–
	NYHA functional class, %
	°Class I	2.6	–	7.4	–
	°Class II	28.9	–	36.8	–
	°Class III	68.5	–	55.8	–
	LVEF≤25, %	53.7	–	59.1	–
	Other comorbidities according to ICD-10a, %
	°Neoplasms, blood (ICD-10; C, D)	14.2	3.2	9.8	–
	°Endocrine, nutritional and metabolic diseases (ICD-10; E)	76.2	8.4	81.7	4.8
	°Nervous system, eye, ear and adnexa (ICD-10; G, H)	21.5	6.4	18.3	16.6
	°Circulatory system (ICD-10; I — excluding chronic heart failure)	76.2	5.3	80.3	11.9
	°Respiratory system (ICD-10; J)	21.4	4.2	29.6	9.5
	°Digestive, skin and subcutaneous system, musculoskeletal system and connective tissue (ICD-10; K, L, M)	23.8	22.1	9.8	19.0
	°Genitourinary system (ICD-10; N)	21.4	–	21.1	–
	Patients with somatic comorbidity, %	97.6	45.3	98.6	45.2
	Somatic comorbidities per patient, M (SD)	4.0 (2.1)	1.2 (0.6)	3.9 (1.8)	1.4 (0.6)

a Diseases are grouped according to the chapters of ICD-10 (International Classification of Diseases 10th Revision); characters represent the respective chapter.

3.2. Symptom pattern for patients with major depressive disorder 

Depression profiles of the major depression patients with and without CHF (n=42 and n=95 respectively) were compared using item-level analyses of covariance for each item of the PHQ-9. Adjusted means, standard deviations, effect sizes, and test statistics are detailed in Table 2. Analyses revealed significant differences with respect to two depression symptoms: Compared to patients without CHF, patients with CHF reported significantly lower scores for depressed mood/hopelessness (PHQ-9, item 2: “Feeling down, depressed or hopeless”) and for worthlessness/feelings of guilt (PHQ-9, item 6: “Feeling bad about yourself — or that you are a failure or have let yourself or your family down”). No significant differences were found between patients with and without CHF for the other depression symptoms, including loss of interest, sleep disturbance, loss of energy, change in appetite, poor concentration, psychomotor agitation/retardation, and suicidal thoughts. For the purpose of illustrating differences in individual symptom severity between major depression patients with and without CHF, Fig. 1 depicts mean differences between both groups and 95% confidence intervals for each of the nine PHQ-9 items.

Table 2. Comparison of CHF patients and non-CHF patients with major depressive disordera

	PHQ-9 items	CHF	Non-CHF	Test statistic
		n=42	n=95
		M (SD)	M (SD)	ESb	F (3,131)c	p
	1. Little interest or pleasure in doing things	2.3 (0.9)	2.5 (0.8)	0.27	1.3	.25
	2. Feeling down, depressed or hopeless	2.2 (0.8)	2.7 (0.7)	0.80	7.7	.006
	3. Trouble falling or staying asleep, or sleeping too much	2.4 (1.1)	2.6 (0.9)	0.23	0.6	.43
	4. Feeling tired or having little energy	2.7 (0.7)	2.7 (0.6)	0.0	0.2	.68
	5. Poor appetite or overeating	2.3 (1.1)	2.1 (1.0)	−0.2	0.4	.55
	6. Feeling bad about yourself — or that you are a failure or have let yourself or your family down	1.4 (1.2)	2.0 (1.1)	0.54	5.7	.019
	7. Trouble concentrating on things, such as reading the newspaper or watching television	1.8 (1.2)	2.2 (1.0)	0.43	2.6	.11
	8. Moving or speaking so slowly that other people could have noticed. Or the opposite — being so fidgety or restless that you have been moving around a lot more than usual	1.7 (1.3)	1.4 (1.2)	0.28	1.2	.28
	9. Thoughts that you would be better off dead, or of hurting yourself in some way	0.9 (1.2)	1.0 (1.0)	0.10	0.02	.88

a Means, standard deviations, effect sizes and p-values are adjusted for age and sex. b Effect size is difference between adjusted means divided by standard deviation of respective item in total sample. c Degrees of freedom in entire patient sample; due to missing data, degrees of freedom vary between df=131 and df=129.

View Large Image Download to PowerPoint

Fig. 1. Differences in severity of individual depression symptoms in patients with major depressive disorder with and without chronic heart failure. Displayed are adjusted mean differences and 95% confidence intervals for individual depression symptoms, as measured with the PHQ-9.

Item-level chi square tests identified significant differences for the very same 2 items (item 2: χ22=8.8, p=.01; item 6: χ32=24.0, p≤.001).

3.3. Symptom pattern for patients with other depressive disorders 

In the group with other depressive disorders, we also compared the depression profiles of patients with and without CHF (n=71 and n=42, respectively). Adjusted means, standard deviations, effect sizes, and test statistics are shown in Table 3. The analyses of covariance completely replicated the results of the patient sample with major depressive disorder: Significant differences in depressive symptom severity were only found for depressed mood/hopelessness (PHQ-9, item 2) and worthlessness/feelings of guilt (PHQ-9, item 6), with the CHF patients showing lower severity scores than the patients without CHF. No significant differences were found for the other 7 depressive symptoms. Mean difference scores with 95% confidence intervals are displayed in Fig. 2.

Table 3. Comparison of CHF patients and non-CHF patients with other depressive disordersa

	PHQ-9 items	CHF	Non-CHF	Test statistic
		n=71	n=42
		M (SD)	M (SD)	ESb	F (3,105)c	p
	1. Little interest or pleasure in doing things	2.0 (0.9)	2.0 (1.0)	0.0	0.04	.84
	2. Feeling down, depressed or hopeless	1.2 (0.9)	2.0 (1.9)	0.89	13.3	≤.001
	3. Trouble falling or staying asleep, or sleeping too much	1.7 (1.1)	1.7 (1.2)	0.0	0.0	.99
	4. Feeling tired or having little energy	1.9 (1.0)	2.1 (1.1)	0.21	0.9	.34
	5. Poor appetite or overeating	0.9 (0.9)	1.0 (0.9)	0.11	0.2	.63
	6. Feeling bad about yourself — or that you are a failure or have let yourself or your family down	0.4 (0.8)	0.9 (0.9)	0.63	4.2	.04
	7. Trouble concentrating on things, such as reading the newspaper or watching television	0.8 (1.0)	1.0 (1.1)	0.21	0.8	.37
	8. Moving or speaking so slowly that other people could have noticed. Or the opposite — being so fidgety or restless that you have been moving around a lot more than usual	0.6 (0.9)	0.9 (0.9)	0.36	2.4	.12
	9. Thoughts that you would be better off dead, or of hurting yourself in some way	0.2 (0.6)	0.2 (0.6)	0.0	0.004	.95

a Means, standard deviations, effect sizes and p-values are adjusted for age and sex. b Effect size is the difference between adjusted means divided by standard deviation of respective item in total sample. c Degrees of freedom in entire patient sample; due to missing data, degrees of freedom vary between df=105 and df=99.

View Large Image Download to PowerPoint

Fig. 2. Differences in severity of individual depression symptoms in patients with other depressive disorders with and without chronic heart failure. Displayed are adjusted mean differences and 95% confidence intervals for individual depression symptoms, as measured with the PHQ-9.

Again, item-level chi square tests did not significantly alter the results (item 2: χ32=16.9, p=.001; item 6: χ32=13.9, p=.003).

4. Discussion 

The major finding from this study is that the profile of depression symptoms in patients with CHF differs from that of patients without CHF. Specific differences in the depression symptoms profiles were found with respect to 2 of the 9 DSM-IV A-criteria for major depressive disorder (American Psychiatric Association, 2000): (a) depressed mood and hopelessness, one of the core depression symptoms, and (b) feelings of worthlessness or guilt, which is an important secondary symptom of depression. While patients with CHF experienced lower levels in these two cognitive–emotional depression symptoms compared to patients without CHF, no differences were found with respect to the somatic symptoms of depression, i.e. sleep disturbance, loss of energy, change in appetite, weak concentration, and psychomotor agitation/retardation. The validity of the findings is supported by the fact that the profile of depression found in CHF patients with major depression was completely replicated in an independent group of CHF patients with other depressive disorders.

The finding that somatic symptoms did not differ between depressed patients with and without CHF is in line with the results of Simon and von Korff (2006), who found only modest differences in the magnitude of somatic symptoms between depressed patients with and without medical comorbidity and concluded that somatic symptoms are not less valid indicators of depression in patients with chronic physical illness. Our results regarding the lower severity of depressed mood and feelings of guilt in patients with CHF are similar to those of an earlier study of patients with and without comorbid physical illness (Clark et al., 1998). Altogether, there is now growing evidence that the features discriminating depressed patients with and without comorbid physical illness are not the somatic depression symptoms but the cognitive–emotional symptoms such as hopelessness and feelings of guilt (Clark et al., 1998).

This finding has several implications for clinical practice, research, and medical teaching: Firstly, it may partially explain the low recognition rate of depression in patients with CHF (O'Connor and Joynt, 2004). Given their specific depression profile, CHF patients might report fewer feelings of depression, hopelessness, worthlessness, and guilt in their medical history, so that clinicians might not consider the presence of a depressive comorbidity. Due to the substantial overlap of somatic symptoms between depressive disorders and CHF, depression may often be dismissed as the somatic symptom burden of heart failure. To improve this situation, medical training should address the fact that depressive disorders may appear in different forms in patients with and without CHF. Secondly, this finding is relevant to the future classification of depressive disorders in DSM-V and ICD-11. Given the similar profiles of somatic symptoms found in depressed patients with and without CHF, as well as in depressed patients with and without comorbid physical illness as described by Simon and von Korff (2006), we agree that a modification of the classification criteria for somatic symptoms specific to patients with medical comorbidity is not necessary. Changes to the classification criteria for cognitive–emotional symptoms of depressive disorders also do not appear to be necessary given that these symptoms constitute a crucial component in the overall picture of depression. Thirdly, it may not be justified to omit somatic symptoms from depression measures, such as the Hospital Anxiety and Depression Scale (HADS) (Bjelland et al., 2002, Zigmond and Snaith, 1983), to avoid overlap with physical symptoms. In fact, psychometric depression measures that include the somatic depression symptoms in accordance with the current classification systems (American Psychiatric Association, 2000, World Health Organization (WHO), 1992) can be used in patient groups with and without CHF without restrictions.

This study did not investigate why patients with CHF experience depression differently from patients without CHF. The finding of a recent interview study, that 38% of depressed medical patients understand their emotional problems as consequence of their physical illness (Löwe et al., 2006), suggests that patients with medical conditions may attribute their depression to their physical illness as an external, non-controllable factor. This external attribution of depression might prevent these patients from self-reproaches, feelings of worthlessness and guilt. However, more research is necessary in order to better understand the reasons for the specific depression profiles in different patient groups.

Several limitations of our study should be noted. Firstly, as expected, our patient samples with and without CHF differed in age and gender. However, this difference appropriately reflects the natural composition of depressed patients with and without CHF in patient populations (Cleland et al., 2003). To account for this difference, all results were controlled for age and gender. Nevertheless the question arises as to whether controlling for these variables in covariate analyses is sufficient in correcting the sample differences, given the potentially insufficient overlap of their distributions. In addition, diagnosis and severity of depression were controlled for by performing separate analyses for patients with major depressive disorder and other depressive disorders, separately. Finally, the patients with and without CHF were recruited in two different clinics, a CHF outpatient clinic and the psychosomatic outpatient clinic of a large Medical University Hospital. Our study population of CHF patients may not be representative for CHF patients recruited in the general population or primary care, respectively, regarding age and etiology of CHF (Lloyd-Jones et al., 2002, Mosterd et al., 1999). Of course, this may also be the case for the depression group. Thus, it is possible that the study results do not only represent the differences in the depression profiles of patients with and without CHF, but also differences in depression profiles of patients presenting in a specialized medical versus a specialized psychosomatic treatment setting. Nevertheless, recent studies have shown that depression prevalence is independent from the etiology of CHF (Faris et al., 2002, Turvey et al., 2002). Based on our findings it appears promising to repeat our analyses in depressed patients with and without CHF from general clinical settings. Because we limited our physically ill sample to patients with documented CHF, we were not able to determine whether other patient groups with physical diseases had depression profiles similar to those of the CHF patients. However, additional studies in well-defined medical patient populations are necessary to determine whether the depression profile, including lower levels of depressed mood and feelings of worthlessness and guilt, is unique to patients with CHF or whether it also applies to other patient groups.

What conclusions can be drawn from the study results for routine clinical practice? The presence of physical symptoms in patients with CHF is almost normal and can of course be the result of heart disease. If a patient does not only describe but complains about his symptom burden; if there is a discrepancy between the objective measures of heart failure severity (left ventricular ejection fraction, natriuretic peptides, etc.), and the symptom complaint; if heart failure remains stable but the symptom complaint increases; or if, on the contrary, the patient does not complain about anything or even hardly talks any more — all of that can be a sign that the patient simply is not well. So, the primary means of diagnosing depressive disorders in CHF patients is to be aware of it and to encourage the patient to report on his subjective well-being. Considering the lack of time in clinical routine, the use of short and easy to handle screening instruments like the HADS (Herrmann, 1997, Zigmond and Snaith, 1983), the PHQ-9 (Holzapfel et al., 2007, Löwe et al., 2004d), or the PHQ-2 (Kroenke et al., 2003, Löwe et al., 2005) may help to successfully complete this important task.

In the case of a diagnosed depression, the question as to therapeutic options arises. Although there is excellent evidence about the effectiveness of pharmacotherapy and psychotherapy in treating depression (Arroll et al., 2005, Hollon et al., 2005, Pampallona et al., 2004, Whooley and Simon, 2000), up to now there have been no treatment guidelines that take into account the special needs of patients with depression and chronic heart failure (Lane et al., 2005). Therefore a lot of work has to be undertaken to improve diagnostic and therapeutic strategies for better treatment of patients with chronic heart failure and depressive comorbidity.