| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Section of General Hospital Psychiatry, Division of Psychological Medicine, Institute of Psychiatry and Guy’s, King’s and St. Thomas’ School of Medicine, King’s College London, London, UK (G.J.R., A.C., M.H.); Section of Neurobiology of Mood Disorders, Division of Psychological Medicine, Institute of Psychiatry, King’s College London, London, UK (G.J.R., A.C.).
Address correspondence and reprint requests to James Rubin, Section of General Hospital Psychiatry (PO62), Weston Education Centre, Cutcombe Road, London SE5 9RJ, UK. E-mail: g.rubin{at}iop.kcl.ac.uk
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
METHODS: One hundred eighty-three patients completed questionnaires before surgery and then 2 days, 3 weeks and 6 months afterward. Multiple regressions were used to assess which of the following factors were important in predicting self-reported postoperative fatigue, controlling for preoperative fatigue and for various demographic and surgical variables: negative mood, history of mood disorder, preoperative expectations of fatigue, preoperative worry or optimism about surgery, preoperative beliefs about the benefits of activity or rest, self-reported postoperative activity, self-reported cardiovascular deconditioning, and availability of social support.
RESULTS: Controlling for demographic and surgical variables and preoperative fatigue, postoperative fatigue showed significant associations with negative mood at each stage of follow-up (p < .001) and was significantly predicted by history of mood disorder at 2 days postoperatively (p = .02). Higher fatigue expectations were self-fulfilling at 3 weeks after surgery (p = .02), whereas preoperative belief in physical activity as being beneficial to recovery predicted reduced fatigue at 6 months (p < .001). Finally, self-reported breathlessness after exercise, an indicator of cardiovascular deconditioning, was also significantly associated with greater fatigue at 6 months (p = .02).
CONCLUSION: The results indicate that psychological processes may well be relevant in the etiology of postoperative fatigue. In particular, the results relating to mood and expectations suggest that somatization may be particularly important in the first few weeks following surgery, whereas cognitive-behavioral factors and cardiovascular deconditioning may be more important in determining later-stage recovery.
Key Words: fatigue, • surgery, • expectations, • somatization, • cognitions.
Abbreviations: ASA status = American Society of Anesthesiologists physical status score;; T1 = preoperative;; T2 = 2 days postoperatively;; T3 = 3 weeks postoperatively;; T4 = 6 months postoperatively;; GHQ-12 = General Health Questionnaire (12 items);; STAI-6 = State Anxiety Inventory (6 items);; IPQ-R = Revised Illness Perception Questionnaire.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Two main theories exist regarding the etiology of postoperative fatigue, one primarily based around physical response to surgery (3,4) and one primarily based around emotional responses (5). The physical theory asserts that postoperative fatigue is mainly the result of decrements in physical fitness, as exemplified by reductions in skeletal muscle and cardiorespiratory functioning. These changes are believed to be caused by the body’s catabolic endocrine-metabolic response to surgery, by reduced nutritional intake and by restricted postoperative activity. Although correlational evidence suggests that postoperative fatigue is related to these three mechanisms, it is notable that nutritional interventions are ineffective in treating the symptom (2) and that some operations such as orthopedic surgery, which produce major endocrine responses, do not seem to cause postoperative fatigue (6). Of the three mechanisms, therefore, changes in activity may be the most important.
After surgery, activity levels can be reduced for several weeks (7,8). As a result, it is likely that cardiovascular fitness will deteriorate (9), requiring a patient to expend more energy for any given action and possibly generating sensations of tiredness. For example, there is good evidence that the fatigue experienced 20 days or more after surgery correlates with objective measures of cardiovascular fitness (3,4,10), although no such correlation is apparent before this (3). The reasons underlying reduced activity after surgery are probably multifactorial and include the impact of fatigue itself, pain during mobilization, negative mood, and the presence of drips and drains (11). Although some of these have already been shown to correlate with increased fatigue (12), another relevant factor that is often overlooked is a patient’s own cognitions. For example, two cues that may be particularly important in regulating activity after surgery are a patient’s interpretation of the medical advice they receive and their interpretation of what their symptoms signify (7). For example, patients often restrict their activity levels in order to avoid or reduce symptoms that they find worrying (1,7), even though such avoidant coping strategies may actually worsen cardiovascular deconditioning and hence delay recovery. Meanwhile, a patient’s general attitude toward the supposed benefits of "rest and recuperation" during illness may predispose some to be more sedentary after surgery than others. Thus, being unwilling to engage in physiotherapy in the first few weeks after orthopedic surgery has been shown to predict higher levels of fatigue 6 months later (13). Finally, the level of social support available to patients may also partly determine their postoperative activity levels. Although greater support has been linked to better emotional outcome after an operation (14), it has also been suggested that having more help with everyday chores might delay mobilization (7). Although two studies that have examined this issue failed to identify any link between postoperative support and fatigue levels (15,16), greater support has been shown to be a risk factor for increased fatigue after viral infection (17).
An alternative, but not mutually exclusive, theory of postoperative fatigue is that suggested by Salmon and Hall (5). According to these researchers, the symptom might best be understood as the result of somatization, whereby patients experience negative mood after surgery but misinterpret it as "fatigue" partly because of the highly medicalized context in which they find themselves and partly because they already expect fatigue and are therefore actively monitoring for it. In support of this, numerous studies have identified strong correlations between negative mood and postoperative fatigue (5), although the direction of causality implied by this is not always made clear. Similarly, it has been noted that expectations of worse recovery after surgery can indeed be self-fulfilling (18), although, again, whether somatization is the mechanism underlying this finding is uncertain.
One of the strengths of the somatization theory is that it offers a good explanation as to why certain operations tend to result in more fatigue than others (5,6). In particular, it is striking that those procedures that patients are normally most willing to undergo also tend to result in the least lethargy. For example, total joint arthroplasty produces little fatigue (6) and also typically presages a welcome return to mobility while being seen by patients as a relatively low risk procedure. Major abdominal surgery, on the other hand, produces large increases in fatigue (6) and is normally seen as a riskier operation with fewer obvious quality-of-life benefits. The perceived risk-benefit tradeoff involved in a procedure may therefore be a crucial determinant of whether a patient subsequently suffers from "fatigue," although other systematic differences between the operative groups can make such interpretations difficult (6).
The prospective cohort study reported here tested the extent to which key variables pertinent to somatization and cognitive-behavioral processes could explain the variation in fatigue after elective inpatient surgery, when important demographic and surgical variables were held constant. For the cognitive-behavioral variables, we tested whether preoperative belief in the supposed efficacy of rest and activity, actual postoperative activity level, cardiovascular deconditioning, and greater availability of social support were associated with higher levels of postoperative fatigue. For the somatization theory, we predicted that greater emotional distress, previous history of mood disorder, preoperative expectations of fatigue, and preoperative beliefs about the risks and benefits of surgery would be associated with greater postoperative fatigue.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Design
This prospective cohort study involved participants completing questionnaires at four time points; preoperatively (T1), 2 days postoperatively (T2), 3 weeks postoperatively, (T3) and 4 months postoperatively (T4).
Measures of Fatigue
Fatigue at each time point was measured using the Chalder Fatigue Scale (19), scored using the "Likert method." This scale has good psychometric properties (19) and has previously been used to good effect in other studies of post-surgical convalescence (20,21).
Measures of Independent Variables
Beliefs about the efficacy of postoperative rest and activity were assessed at T1 using eight items adapted from a previous study (22). These asked how patients thought physical activity or rest would affect their recovery in the first 3 weeks or 6 months after surgery, if in pain after surgery, and if fatigued after surgery. A ninth item assessed whether patients believed fatigue-inducing activity should be avoided if a person is already feeling tired. Results for these items were combined into two scales relating to belief about rest and belief about activity. The decision to do this was taken a priori and guided by a principal components analysis of the relevant items. This revealed two components with an Eigen value >1, accounting for 52.5% of the variance of the original items. Following Varimax rotation, all rest-related items loaded most heavily onto one component (Cronbach’s 
= 0.73), whereas all activity-related items loaded most heavily onto the other (Cronbach’s = 0.74); belief about the benefit of fatigue avoidance loaded onto neither component and was dropped from the analysis.
Actual postoperative activity levels were assessed using items asking whether any exercise and any sweat-inducing exercise had been performed in the past 3 (T3) or 4 (T4) weeks. Cardiovascular fitness was assessed by a single item at T4 which asked whether participants now got excessively breathless after exercise, an item previously shown to predict the development of chronic fatigue in primary care (17). The availability of social support was assessed at T1 using an abbreviated version of the Significant Others Scale (23).
Emotional distress at T1, T3, and T4 was assessed using the General Health Questionnaire (24) and a six-item short form of the State-Trait Anxiety Inventory (STAI-6) (25). Because of its brevity, only the STAI-6 was used at T2. Previous history of mood disorder was assessed at T1 through a single item asking, "Have you ever been to your GP or any other doctor because of an emotional problem."
Preoperative recovery expectations were assessed using eight items. The first asked patients to record how severe they expected any postoperative fatigue would be in the first 3 weeks after surgery using a seven-point Likert scale. Patients were also asked open-ended questions about how long they expected any postoperative fatigue would last for and when they expected to be able to return to their normal daily routine. Four items adapted from the Work and Social Adjustment Scale (26) were also presented, amended so as to ask how impaired patients believed they would be in terms of work, home management, social leisure activities, and private leisure activities in the first 3 weeks after surgery. Finally, a single seven-point Likert item was included which asked patients to report how severe their postoperative fatigue was the last time they had surgery; if a patient had no previous personal experience of surgery he or she was instead asked to report how much fatigue any close friend or relative had had. As with the belief about rest items, results from the expectation items were combined into two scales. Again, a principal components analysis confirmed the presence of two components with Eigen values >1 accounting for 64.8% of the variance in the initial items. After Varimax rotation, items derived from the WSAS loaded most heavily onto one component ("expected impairment," Cronbach’s = 0.87), while those relating to expected fatigue duration and severity and to previous experience of postoperative fatigue loaded most heavily onto the second ("expected fatigue," Cronbach’s = 0.59). Expected duration of convalescence did not load onto either and was not analyzed further.
Beliefs about the risks of surgery were assessed at T1 using a single seven-point Likert scale which asked, "how worried are you about having your operation." Beliefs about the possible benefits of surgery were assessed using the treatment control subscale of the Revised Illness Perceptions Questionnaire (IPQ-R) (27), a scale which assesses how optimistic a patient is that their treatment will be effective in improving their condition.
Measures of Demographic and Surgical Variables
Key demographic and surgical data were collected for inclusion in the analyses as variables that might need to be controlled for. The demographic data measured in the T1 questionnaire consisted of age, sex, ethnicity (white vs. other), socioeconomic status (manual vs nonmanual work), and educational level (educated past age 16 vs. not). The clinical data, which were extracted from surgical notes at T4, consisted of type of operation (minor surgery, major abdominal or vascular surgery, gynaecological surgery or cardiac surgery), American Society of Anesthesiologists physical status score (ASA status; a subjective global health rating usually completed by the patient’s anesthetist), duration of anesthesia, whether a blood transfusion was given, mean daily intake of opioid analgesia (in morphine equivalent dosage) (28) and whether the patient experienced a surgical complication.
Procedure
Eligible patients were invited to take part in a study looking at "why some patients recover faster from surgery than others." Those who provided written consent were then asked to complete the T1 questionnaire at home and return it by mail. Two days after surgery, patients were visited in the hospital and asked to complete their T2 questionnaire. Three weeks and 6 months later, patients were sent the T3 and T4 questionnaires and asked to return them in the envelopes provided. Patients who did not return these questionnaires received two written reminders and a follow-up telephone call.
Ethics
This research was conducted in accordance with the Declaration of Helsinki. Ethical approval for the study was given by the King’s College Hospital Research Ethics Committee.
Analysis
Multiple regressions were used to assess the importance of the independent variables in predicting fatigue at T2, T3, and T4. For each of these three time-points, two regression models were developed. In the first model, all the surgical and demographic variables were entered as independent variables, as was preoperative fatigue; postoperative fatigue was used as the dependent variable. Any independent variables which were not significant (p > .1) were then removed and the regression was rerun. For the second model, we then added in relevant preoperative or contemporaneous psychological variables as additional independent variables. Again, those which were not significant (p > .1) were removed and the regression was run again in order to determine the best fit.
For the purposes of the regressions, type of surgery was transformed into three dummy variables, with minor surgery taken as the reference category.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
The results of the multiple regressions used to predict fatigue at T2, T3, and T4 are shown in Tables 2, 3 and 4
. Each of the regressions for the second models showed a significant fit to the data (p < .001) and accounted for 20.7% of the variance in fatigue at T2, 36.7% at T3, and 69.3% at T4. Controlling for preoperative fatigue and for relevant surgical and demographic variables, anxiety or emotional distress showed highly significant associations with fatigue at each postoperative time point (p < .001). History of mood disorder was also positively associated with fatigue at T2 (p = .02). Fatigue at T3 was significantly predicted by greater preoperative expectations of fatigue (p = .02). Finally, at T4, having a firmer belief in the efficacy of postoperative physical activity significantly predicted reduced fatigue (p < .001), whereas being excessively breathless at T4 was significantly associated with greater fatigue (p = .02).
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
By 6 months postoperatively, a different set of factors seems to have become more relevant. As with previous studies (3,4,10), our results show that late-stage postoperative fatigue seems to be associated with cardiovascular deconditioning. Again, it is difficult to make causal assumptions based on just this one finding; fatigue reduces activity levels and fitness just as reduced fitness can produce sensations of fatigue. Yet the additional finding that preoperative beliefs about physical activity predict fatigue levels at 6 months does suggest that reduced activity, caused in part by cognitive factors, is partly responsible for subjective fatigue at this time point. Furthermore, the finding that almost 70% of the variability in fatigue 6 months after surgery could be accounted for by preoperative fatigue, sex, cognitions about activity, cardiovascular fitness, and emotional distress is striking: surgical factors would seem to have very little to do with long-term "postoperative" fatigue.
The other somatization and cognitive-behavioral variables that were assessed in this study showed no significant relationship to postoperative fatigue. For example, none of the measures of postoperative activity were associated with fatigue, a surprising finding, given that associations were identified for both self-reported cardiovascular deconditioning and activity-related cognitions. It may be that this lack of association relates more to the poor quality of the self-report measures that were used than to an absence of any effect; certainly, more in-depth qualitative studies of activity after surgery do suggest that a correlation with fatigue exists (7). Future research should ideally make use of digital activity monitors to examine this issue further. Although this has been attempted in one earlier study which failed to show any association (8), possible demand characteristics in that study and the development of unobtrusive activity monitors now make the issue worth revisiting.
Availability of social support was also unrelated to fatigue severity. Therefore, although it has been suggested that "the presence of more helping hands" can delay recovery (29), this study adds to the consensus that social support is rarely a causal factor in the presence of postoperative fatigue (15,16).
Finally, no evidence was found to link patient perceptions regarding the risks or benefits of surgery to increased fatigue levels. It may be that the assumption that these variables are the key determinants of postoperative emotional distress, and hence fatigue, is oversimplified. Instead, a range of factors is likely to be important in determining distress, including personality, loss of control, loneliness, an inability to fulfill social obligations, and the presence of unpleasant symptoms. A more in-depth assessment of such factors might still reveal that determinants of negative mood also predict postoperative fatigue.
The results reported here were obtained from a relatively large sample with a good response rate and using a psychometrically validated fatigue measure. Nevertheless, the correlational nature of some of the results presents difficulties with their interpretation. Good-quality randomized controlled trials should now be used to replicate and extend these findings and to determine if they have any clinical application (2). For example, it might be interesting to see whether postoperative antidepressants or mood-enhancing psychotherapies are effective in reducing the severity of postoperative fatigue. Alternatively, a brief preoperative cognitive-behavioral intervention aimed at altering patient cognitions about the efficacy and safety of postoperative activity might be useful, especially as a leaflet-based approach using these principles has recently been shown to be a cheap yet effective way of reducing fatigue after glandular fever (30).
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
This research was supported by a Medical Research Council Studentship awarded to James Rubin.
Received for publication January 23, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  G. J. Rubin, M. Hotopf, A. Papadopoulos, and A. Cleare Salivary Cortisol as a Predictor of Postoperative Fatigue Psychosom Med, May 1, 2005; 67(3): 441 - 447. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
