 |
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
|
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
*
Department of Medicine, George Washington University Medical Center,
Washington, D.C.
Department of Psychiatry and Behavioral Sciences, George Washington
University Medical Center, Washington, D.C.
School of Public Health, George Washington University Medical Center,
Washington, D.C.
§
Department of Psychology, George Washington University, Washington,
D.C.
||
Department of Human Development and Psychoeducational Studies, Howard
University Medical Center, Washington, D.C.
¶
Department of Medicine, Howard University Medical Center, Washington,
D.C.
#
Section on Women's Health, Developmental Endocrinology Branch, National
Institute of Child Health and Development, National Institutes of Health,
Bethesda Maryland
**
Department of Medicine, Washington Hospital Center, Washington,
D.C.
Correspondence to Dr. Paul L. Kimmel, Division of Renal Diseases and Hypertension, Department of Medicine, George Washington University Medical Center, 2150 Pennsylvania Avenue, N.W., Washington, DC 20037. Phone: 202-994-4244; Fax: 202-994-2972.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Previously, we reported differences in the compliance behaviors of male and female HD patients (17), and that compliance is associated with enhanced survival (7). Recently, clinical changes following stimulation of synthesis or administration of interleukin-1 (IL-1), such as anorexia, fatigue, somnolence, and lethargy, have been compared with the psychobiologic characteristics of depression (21), and patients with depression have been noted to have increased circulating cytokine levels (21). Worsened perceptions of intrusive illness effects and higher levels of circulating proinflammatory cytokines were associated with increased mortality in a prospective evaluation of HD patients (7,22). No study has investigated the relationship between plasma levels of cytokines and ß-endorphins (as markers of immune dysfunction and stress), and indices of dyadic satisfaction and depression in patients with chronic disease, or associations between mortality and dyadic satisfaction in ESRD patients treated with HD. We present further analyses to assess whether the perception of dyadic satisfaction and conflict was associated with differential survival in men and women with ESRD who were treated with HD and involved in dyadic relationships.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Psychosocial variables were assessed at study entry and at 6-mo intervals thereafter for 1 yr to test their stability. The criterion for a stable dyadic relationship was one that had lasted for 6 mo or more, as used by other investigators (23). Of the 295 patients who consented to participate in the larger study (recruitment rate 60.8%) (7), 174 (59%) were involved in a dyadic relationship of greater than 6 mo duration.
All measures except dyadic adjustment parameters have been presented and discussed previously (7,22), and are only briefly described here. Medical risk factors included a disease severity coefficient (including a diabetes code), the mean of 3 sequential months' serum albumin concentrations (S[Alb]), Kt/V values, protein catabolic rates (PCR) obtained after enrollment, and dialyzer type (unmodified cellulose versus modified cellulose or synthetic) (7,22). Psychosocial measures included the Beck Depression Inventory (BDI) and its cognitive depression subscale (5,7,24,25) to assess depression, the Illness Effects Questionnaire (IEQ) as a measure of perception of illness effects (7,26,27,28), and the Multidimensional Scale of Perceived Social Support (MSP) (29) to assess overall perceived support (7,29). Behavioral compliance was assessed for 90 d after study entry as Percent Time Compliance (% COMP), a measure of treatment session shortening, Percent Attendance (% ATTN), a measure of skipping behavior, and Total Time Compliance (% TCOMP), a measure of the total amount of prescribed treatment time missed in both attended and unattended sessions (7,17).
Blood samples were obtained at the regular monthly laboratory evaluation immediately after study enrollment, before initiating HD (22). Plasma interleukin-1 (IL-1) concentration, as a marker of immune dysfunction, was measured by a chemiluminescence-enhanced capture enzyme-linked immunosorbent assay (22,30,31). Plasma ß-endorphin concentration, a stress marker, was measured by high-performance immunoaffinity chromatography, as described previously (32). Intrassay and interassay coefficients of variation for IL-1 and ß-endorphin measurements ranged from 2.18 to 4.59% (22). Eleven male and 17 female healthy African-American subjects underwent evaluations of circulating IL-1 and ß-endorphin levels, as described previously (22). Their mean age was similar to the HD patients. There was no difference in level of circulating IL-1 or ß-endorphin between the healthy men and women (22). The mean IL-1 and ß-endorphin concentration was 3.5 and approximately 10-fold higher, respectively (P = 0.0001), in HD patients compared with the control group.
The Dyadic Satisfaction Subscale of the Dyadic Adjustment Scale (DAS), an independently valid and reliable measure of relationship satisfaction and adjustment, assessed patients' satisfaction with their marital or partner relationships (33,34). Unmarried subjects were considered as functioning in a dyad if the relationship had been a stable one for more than 6 mo (23). The DAS scores responses to questions such as, "Do you confide in your mate?" and "Rate your level of happiness with this relationship." Higher scores indicate happier relationships. Subscale items are categorized as positive or negative. The DAS positivity score was calculated as the mean of the six positive DAS items. Higher scores reflect positive feelings about the relationship. The DAS negativity score was calculated using the mean of the four negative items of the DAS. Higher DAS negativity scores reflect increased perception of dyadic conflict.
Statistical Analyses
Correlations among selected continuous variables were assessed using
Pearson coefficients, and Spearman rank order coefficients for measures with
skewed distributions (behavioral compliance measures and IL-1 and
ß-endorphin concentrations)
(7,17,22,28).
The correlations between age, disease severity, serum albumin, plasma IL-1 and
ß-endorphin levels, and psychosocial and DAS variables were of particular
interest. Differences between groups were assessed by unpaired t
tests and 
2 analysis. Differences between correlations were
assessed visually, and subsequently the difference between correlation
coefficients in men and women was assessed using the Fisher Z Prime test
(35).
Survival time for each patient was determined as described previously (7,22). Cox proportional hazards regression was used to predict mortality hazard. Preliminary equations were calculated for age at study entry, gender, severity coefficient, S[Alb], PCR, Kt/V, and dialyzer type. After the results of these preliminary analyses, multivariable Cox regression analyses were performed in which the relationship between ß-endorphin and IL-1 levels, compliance indicators, psychosocial and dyadic variables, and survival were examined while simultaneously controlling for the effects of variation in levels of the medical risk factors (age, severity coefficient, S[Alb], dialyzer type, and site) that were significantly associated with survival. Relative risks represent the expected change in mortality risk associated with a 1 SD increase in the predictor variable, except for dialyzer type and gender, allowing comparison of the effects of changes in levels of several risk factors. Analyses were performed using PROC PHREG in SAS 6.12 (SAS Institute, Cary, NC). The alpha level of tests of survival and group differences was 0.05. Data are presented as mean ± SD.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
One hundred six of the 174 patients (60.9%) were married, while 68 of the subjects participated in stable dyadic relationships. There was no statistically significant difference between the proportion of patients functioning in dyadic relationships who were married or unmarried. There was no statistically significant difference between the proportion of married or unmarried patients involved in dyadic relationships who were men or women, or who had diabetes mellitus.
Initial cytokine and neuropeptide data were missing in 23.4% of patients because of study enrollment after final assessments, transfers to other units before assessments, lack of proper identification or hemolysis of specimens, or refused or missed phlebotomies. There was no statistically significant difference between the mean level of circulating IL-1 or ß-endorphin between men and women in the study or between patients with and without diabetes. The only differences between the genders in mean medical, dialytic, cytokine, neuropeptide, psychosocial, or nutritional parameters are shown in Table 2. The women in the study were younger, had a lower level of medical illness severity, and a lower mean S[Alb] than the men. These variables were controlled in survival analyses (see below).
|
The mean BDI score of the patients in the study was in the range of mild depression (Table 1) (24). The patients' mean MSP score was similar to normative samples, and the mean IEQ score was comparable to scores for general medical inpatients and outpatients (7). Behavioral compliance was comparable to that reported in U.S. dialysis populations (7,36). There was no statistically significant difference between the mean values of depression, social support, or perception of illness effects scores of men and women.
The mean total DAS score was 37.8 ± 6.9, similar to the norm mean of 40.5 for married individuals (33). Patients' individual baseline, 6-mo, and 1-yr DAS, dyadic negativity, and dyadic positivity scores correlated highly over time (range r values, 0.67 to 0.85; all P = 0.0001), indicating the dyadic adjustment scales measure enduring characteristics of the relationship in this population. Women had a higher mean DAS negativity score, denoting higher levels of perceived dyadic conflict (Table 2). There was no statistically significant difference in mean DAS or DAS positivity scores between the genders, or between patients with and without diabetes. There was no statistically significant difference in the proportion of patients with and without diabetes mellitus who scored above the median of DAS or DAS positivity scores. There was no statistically significant difference between mean DAS or DAS positivity scores in the groups of patients functioning in dyadic relationships who were married or unmarried.
Older HD patients had lower levels of DAS negativity (r = -0.20, P = 0.01). No dyadic adjustment score correlated with any behavioral compliance measure. DAS scores correlated with ß-endorphin levels (r = -0.16, P = 0.05) in the 174 patients, but there was no correlation of IL-1 concentration and level of any dyadic adjustment parameter.
Table 3 presents the correlations between the psychosocial and the medical, cytokine, and neuropeptide variables that were different between men and women by visual inspection. In women, increased depression scores correlated with higher severity coefficient scores, and with increased levels of ß-endorphin and IL-1. Higher IEQ scores correlated with higher severity coefficient scores and IL-1 levels in women. These parameters were not correlated in men. The correlation coefficients for BDI and the severity coefficient were significantly different in the men and the women (Fisher Z Prime test).
|
Visual examination of the relationships between dyadic adjustment indices and medical risk factors, neuropeptide and cytokine levels, and psychosocial factors also revealed divergent relationships in men and women (Table 4). In men, the only medical risk factor that correlated with dyadic adjustment parameters was increased age, with decreased perception of dyadic relationship conflict. In women, higher % ATTN was correlated with higher dyadic satisfaction, and higher perception of dyadic relationship conflict was correlated with lower levels of Kt/V.
|
Furthermore, significant correlations were found between higher IEQ scores, higher level of depression and lower MSP, and decreased dyadic satisfaction and higher perception of dyadic relationship conflict in women (Table 4). In women, greater perception of dyadic conflict and lower levels of dyadic satisfaction were correlated with higher levels of ß-endorphin. Despite the magnitude of the correlations between IL-1 levels and levels of dyadic satisfaction (r = 0.25) and perceived dyadic relationship conflict (r = 0.22) in women, they were not significant, due to the small number of subjects. In contrast, in men, only levels of depression and social support correlated with perceptions of dyadic adjustment.
Only the correlation coefficients between DAS and DAS negativity scores and IEQ scores, and DAS and % ATTN in women, and between the severity coefficient and BDI score in women were significantly different between the genders (Fisher Z Prime test) (Tables 3 and 4).
Mean and median follow-up times from study entry for the population were 32.2 ± 11.4 and 36.8 mo. A total of 123 patients survived and 51 (nine women [22.5%] and 42 men [31.3%]) died during the study (P = 0.28). A Cox regression analysis in the 174 patients confirmed increased mortality risk for increased age and severity coefficient (Table 5). Higher IL-1 and ß-endorphin levels were associated with increased risk, a finding also noted in the subpopulations of men and women. Improved shortening, skipping, and overall behavioral compliance were associated with reductions in mortality risk. There was no association between gender, level of S[Alb], Kt/V, or depression and survival in the population. In contrast, increased social support and decreased IEQ scores were associated with decreased mortality risk.
|
A 1 SD increase in DAS (denoting greater dyadic satisfaction) was associated with a 29% decrease in relative mortality risk. Although the patients' assessment of the positive quality of the relationship was not associated with outcome, a 1 SD increase in the negativity subscore of the DAS score (denoting greater dyadic conflict) was associated with a 46% increased mortality risk.
Among women, higher S[Alb] was associated with a decreased mortality risk, and improved % TCOMP was associated with mortality risk reduction (Table 5). A 1 SD increase in DAS was associated with a 50% decrease in mortality risk, while a decrease in the negativity subscore of the DAS was associated with a 60% decrease in the mortality risk (Table 5, Figure 1). In contrast, the dyadic adjustment parameters were not significantly associated with mortality risk in men (Table 5, Figure 1). We then tested whether the Cox regression estimates differed for men and women. After applying the cross-product method for generating interaction effects, none of the interactions with gender was statistically significant.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Higher perception of dyadic satisfaction was also associated with a decreased mortality risk, while greater perception of dyadic conflict was associated with a 46% increased relative death risk. An unsatisfactory dyadic relationship may place a member of the couple at greater risk for health and psychologic problems, because the relationship may be the source of stress, while at the same time preventing the attainment of social support outside the context of the dyad (37,38). Poorer quality of a relationship has been associated with greater levels of depression and poorer immune function in married women (20). Our data support this view. Although the recruitment of women was hampered by the small population of women treated for ESRD at the Veterans Affairs Hospital, when subsets were analyzed by gender, higher levels of perceived dyadic satisfaction and lower levels of perception of dyadic conflict were significantly associated with enhanced survival in the small group of women in the study. This was not true in the larger subset of men. This result must be viewed with caution because differences in the level of association were not different between men and women. Still, the clinically meaningful levels of mortality prediction among women suggest that the quality of a dyadic relationship may be a very important risk factor for this group.
Women reported higher levels of conflict in dyadic relationships than the men, and dyadic satisfaction for women was not correlated with age, comorbidity, or S[Alb]. There were also gender differences in the correlation of psychosocial variables with dyadic satisfaction and distress. For women, dyadic satisfaction was associated with perception of the intrusiveness of illness, depressive symptoms, and perceived social support. Only women demonstrated an association between increased dyadic satisfaction and behavioral compliance. In addition to psychosocial differences, variation in the levels of IL-1 and ß-endorphin, which have been associated with mortality in previous studies by our group (22,39), were associated with differential dyadic satisfaction and conflict only in women. Correlations may be lower among the men in some cases because of the lesser variability of DAS parameters in this group. Thus, for women, dyadic satisfaction appears to relate to a complex admixture of neuroimmunologic and psychosocial adjustment factors.
Depressed patients have higher circulating levels of IL-1 (21,40) and other acute-phase reactants (21,41) compared with control subjects, and several studies have suggested that cytokines might be trait markers for depression (21). However, the effects of increased production of IL-1 have clinical characteristics similar to the uremic syndrome (42,43). We previously highlighted the similarity of both of the symptoms of depression (5) and abnormal cytokine regulation (22,43), such as fatigue, and appetite and sleep disturbances, to those of uremia. Cytokine dysregulation might be similar in patients with depression and ESRD, providing a link between the pathogenesis of symptoms in the two conditions.
Studies have reported a relationship between marital discord and negative affect and abnormal immune function in members of a dyad (13,44). For women, there appears to be a relationship between psychologic and immune factors, because depression and perception of the intrusiveness of illness effects were correlated with increased levels of IL-1 and ß-endorphin, supporting the notion that emotional status and immune function are more closely linked in women than in men. Such findings have been noted in recent studies of dyadic conflict in Caucasian newlyweds (23) and older married adults (19) in whom reactive variation in neurohumoral and immunologic markers was correlated with psychologic parameters only in women (19,20,23).
The failure of levels of a stress mediator to return to normal after a challenge constitutes an abnormality of the allostatic system of stress responses (45). ß-endorphin levels, which may modulate immune function through immune cell receptors (46,47), are increased in HD patients (48), as are cytokines (22,49), although the mechanisms underlying the dysregulation are not completely understood. In women, increased levels of ß-endorphin correlated with high levels of dyadic conflict and depression. Strong negative emotion, such as perception of dyadic conflict in women, may be a particularly potent stressor, causing physiologic arousal of the autonomic nervous system. The intercorrelations between perception of dyadic conflict, perception of worsened depression, illness intrusiveness, and higher levels of IL-1 and ß-endorphin in women HD patients suggest an association between emotional and cognitive function and dysregulated levels of immunologic and neurohumoral markers. Poorer outcomes may follow in patients who cannot downregulate circulating levels of stress hormones because of renal dysfunction. We did not, however, determine the biologic activity of the plasma neuroimmunologic markers measured.
The finding that higher dyadic satisfaction and lower levels of dyadic conflict among women are associated with a protective effect (Figure 1), compared with women with low dyadic satisfaction and high conflict and men with varying levels of dyadic satisfaction and conflict may also in part be a function of the ethnic background of the vast majority of study participants. African-American women are reported to particularly value positive family interactions and strong family ties (50,51). Positive dyadic and family relationships (51) may provide a unique type of social support and approach to dealing with chronic illness that confers a survival advantage to African-American women. In male HD patients, although immune parameters and compliance are predictors of survival, the psychosocial factors that influence them remain unidentified. Previously, in a similar patient population, we determined that African-American women HD patients, functioning in complex family situations, had higher mortality than women who lived alone or only with a dyadic partner (52). This differential survival may be a result of the stressors present and acting in a population at risk from a mortal illness, when faced with a concomitant increased number of household management duties, and familial roles and responsibilities. Similar findings were also reported by Reiss and colleagues, who found unexpectedly higher mortality in patient members of more cohesive families, largely derived from an urban, African-American ESRD population (53).
Support within the dyad may also change differentially by gender with the onset of ESRD. In couples in which the woman developed ESRD, in a Canadian population of patients, family supportiveness declined after the onset of illness, while no change in perceived support was noted for the male ESRD patients (54). Thus, the decline in relationship support may also be related to high dyadic conflict and its effects among women treated with HD.
The association of greater dyadic satisfaction in women with improved survival does not necessarily suggest causality. Other, unknown mediating factors may be associated with both the limited ability to form or participate successfully in dyadic relationships and medical factors that affect outcome. Poor outcome in women with high perceived levels of marital distress may be mediated by specific immune responses, through social interactions or through compliance with the prescribed dialysis regimen. These studies must be extended both to larger samples of men and women to determine whether gender differentially affects the relationship of perception of dyadic satisfaction and conflict and mortality and to more diverse populations. African-American women with ESRD treated with HD who are satisfied with their relationships, however, appear to be a population at special benefit. Therefore, women who report high dyadic dissatisfaction and conflict may be amenable to therapeutic interventions, which could in turn improve outcomes.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. Holt-Lunstad, W. A. Birmingham, and K. C. Light Influence of a "Warm Touch" Support Enhancement Intervention Among Married Couples on Ambulatory Blood Pressure, Oxytocin, Alpha Amylase, and Cortisol Psychosom Med, November 1, 2008; 70(9): 976 - 985. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. L. Weihs, T. M. Enright, and S. J. Simmens Close Relationships and Emotional Processing Predict Decreased Mortality in Women with Breast Cancer: Preliminary Evidence Psychosom Med, January 1, 2008; 70(1): 117 - 124. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Cukor, S. D. Cohen, R. A. Peterson, and P. L. Kimmel Psychosocial Aspects of Chronic Disease: ESRD as a Paradigmatic Illness J. Am. Soc. Nephrol., December 1, 2007; 18(12): 3042 - 3055. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. K. Kiecolt-Glaser, T. J. Loving, J. R. Stowell, W. B. Malarkey, S. Lemeshow, S. L. Dickinson, and R. Glaser Hostile Marital Interactions, Proinflammatory Cytokine Production, and Wound Healing Arch Gen Psychiatry, December 1, 2005; 62(12): 1377 - 1384. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Chesla, L. Fisher, J. T. Mullan, M. M. Skaff, P. Gardiner, K. Chun, and R. Kanter Family and Disease Management in African-American Patients With Type 2 Diabetes Diabetes Care, December 1, 2004; 27(12): 2850 - 2855. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Kalantar-Zadeh, J. D. Kopple, G. Block, and M. H. Humphreys Association Among SF36 Quality of Life Measures and Nutrition, Hospitalization, and Mortality in Hemodialysis J. Am. Soc. Nephrol., December 1, 2001; 12(12): 2797 - 2806. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
HOME
CURRENT ISSUE
ARCHIVES
JASN Express
ONLINE SUBMISSION
AUTHOR INFO
EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP |
