Volume 40, Issue 2 , Pages 256-265, August 2010
Using Vital Sign Flow Sheets Can Help to Identify Neoplastic Fever and Other Possible Causes in Oncology Patients: A Retrospective Observational Study
Article Outline
Abstract
Context
It is important to determine the etiology of fever in cancer patients. Such patients often undergo extensive laboratory and radiographic investigations and prolonged anti-infective therapy that are time- and resource- consuming, risk drug toxicity, and postpone systemic chemotherapy.
Objectives
To investigate neoplastic fever (NF) patterns from vital sign flow sheets.
Methods
Between September 1997 and February 2009, data on 150 consecutive hospitalized patients with advanced or metastatic solid tumors documented to have NF were retrospectively collected. Sixty patients with sepsis were used as a comparison group.
Results
All patients with NF demonstrated intermittent fever patterns. Peak body temperature was 39.0
±0.6°C (38.0–40.8°C). Baseline pulse rates in 139 (93%) patients showed no increase except during febrile periods. The remaining 11 (7%) patients had transiently elevated baseline pulse rates at the time of cessation of postchemotherapy dexamethasone. Once-daily fever spike patterns occurred in 108 (72%) patients. Fever spikes were most commonly found at 9 am (42%) and 5 pm (37%). Twice-daily fever spike patterns were noted in the 42 (28%) remaining patients. In the comparison group, baseline pulse rate elevated in all patients during febrile periods and 20 (33%) showed intermittent fever patterns.
Conclusion
We conclude that the NF pattern is characterized by intermittent fever without an obvious increase in baseline pulse rate except during febrile periods. Knowing NF patterns from vital sign flow sheets can help identify NF and other possible causes of fever in oncology patients.
Key Words: Neoplastic fever, vital sign flow sheets, intermittent fever, pulse rate, oncology patients
Introduction
Neoplastic fever (NF) is a paraneoplastic syndrome that is caused by cancer itself.1, 2, 3 However, fever in cancer patients is often the result of infection. It is important to determine the etiology of fever in cancer patients.1, 2, 3, 4 Patients often undergo extensive laboratory and radiographic investigations and prolonged anti-infective therapies that are time- and resource-consuming, increase risk of drug toxicity, and lead to the postponement of systemic chemotherapy.1 No markers of inflammation, radiographic tests, or the naproxen test can confirm the diagnosis of NF.4 C-reactive protein and procalcitonin have been used in differentiating infections from NF;5, 6 however, studies have failed to show that they can discriminate between infective and noninfective causes of fever.7
Antibiotics and standard antipyretic drugs such as acetaminophen (paracetamol) are usually ineffective in patients with NF. A negative blood culture and fever unresponsive to antibiotics cannot exclude an uncontrolled bacterial infection. Nonsteroidal anti-inflammatory drugs (NSAIDs), such as naproxen or indomethacin, are effective in ameliorating the NF.8, 9, 10, 11, 12 The naproxen test is a clinical tool in the differential diagnosis of fever of undetermined origin in cancer patients.1, 2, 8, 9, 10, 11, 12, 13 The standard test of naproxen dose is 500mg every 12 hours for two days. A small portion of patients with NF fail to defervesce on naproxen, and it has been reported that the naproxen test lacks value in the diagnosis of fever because of malignancy.1, 14 Moreover, in some patients, naproxen cannot be administered by mouth because of gastrointestinal obstructions or gastrointestinal irritation or bleeding.
NF usually exhibits intermittent fever patterns.15 Intermittent fever is characterized by intermittent episodes over several days, and alternate with afebrile periods. But intermittent fever also can be of an infective origin.16, 17
The goal of this study was to investigate NF patterns from data recorded on vital sign flow sheets. We hoped that fever patterns from vital sign flow sheets can help us to identify early the etiology of fever in cancer patients.
Methods
Patients
Between September 1997 and February 2009, data on 150 hospitalized patients with advanced or metastatic solid tumor documented to have NF were collected retrospectively at Chang-Gung Memorial Hospital. The criteria of suspected NF included the following: 1) temperature exceeding 38°C at least once, 2) no sign of infection in clinical findings and radiological examinations, 3) negative blood and other cultures, 4) lack of response of the fever for at least five days, and 5) adequate antibiotic treatment for at least seven days. During febrile periods, body temperature (measured with the use of an ear thermometer), blood pressure, pulse rate, and respiratory rate were observed from vital sign flow sheets. After at least five days of persistent fever despite adequate antibiotic treatment, without a positive culture or infection source and with stable vital signs, the fever was regarded as malignancy related.
When NF was suspected, patients were tried on naproxen. The dosage of naproxen ranged from 125 to 250mg every 12 hours. Acetaminophen, tramadol, codeine, morphine, and transdermal fentanyl were common drugs given to control pain in these patients.
The definition of NF needs to be rigorous. Those who died, were lost to follow-up, or had an infection within two weeks of NF being suspected were excluded from this analysis. Drug fever was difficult to diagnosis.18, 19 There was no evidence that any drugs taken by patients in this study could induce fever. If patients after being suspected of having NF received chemotherapy without any untoward consequences, the origin of their fever from infection was ruled out.
The characteristics of NF were described as including the highest peak body temperature, usually the daily peak temperature; heart rate alteration; presence or absence of chills; association with other paraneoplastic syndromes; and white blood counts and their band forms. Leukemoid reaction was defined as the elevation of the peripheral count to more than 20,000/μL without evidence of infection or leukemia. Hypercalcemia was defined as a serum calcium level more than 11mg/dL. A significant elevation in baseline pulse rate was defined as a 10% sustained increase in baseline pulse rate during febrile periods.
The patterns of NF were investigated, including the daily fever spike pattern and its predominant time. Once-daily fever spike patterns included either a pure single-daily fever spike or a single-daily fever spike mingling with a few twice-daily fever spikes during febrile periods. Twice-daily fever spike patterns were defined as more than 50% of daily fever spikes being double spikes during febrile periods.
Comparison Groups
For the purpose of identifying fever etiology early in cancer patients and investigating the difference of fever patterns between the NF and the infection groups, cancer patients with the complication of sepsis were retrospectively collected as a comparison group. Their fever patterns were studied between January 2005 and February 2009. They included 60 oncology patients with septicemia who had a positive blood culture. The highest peak body temperature, usually the daily peak temperature; heart rate alteration; and white blood counts and their band forms were retrospectively collected from chart records and vital sign flow sheets.
Statistical Methods
Vital sign flow sheets were obtained from the hospital computer system. We used the Chi-squared and Student's t-tests to detect the differences between subgroups and considered a P<0.05 as an indication of statistical significance. The duration of survival was calculated from the time of NF suspicion to death. But death occurred within two weeks after NF suspicions were excluded. The survival curves were determined using Kaplan-Meier methods.
Results
A total of 162 consecutive cancer patients had retrospective data collected for evaluation of the presence of NF. The causes of exclusion (n=12) were as follows: death (n=7), lost to follow-up (n=4), and occurrence of infection (n=1) within two weeks after NF suspicion. The remaining 150 patients' characteristics are shown in Table 1. There were 104 men and 46 women, ranging in age from 20 to 84 years (median 56 years). NF occurred in various malignancies; 24 (16%) of the cases exhibited fever before cancer was diagnosed and were either metastatic or locoregionally advanced cases. Seventy-two (48%) patients had an Eastern Cooperative Oncology Group performance status of 2 or greater.
Table 1. Characteristics of NF According to Patients' Characteristics
| Characteristics | Cases | % |
|---|---|---|
| Age, years | ||
| Median (range) | 56 (20–84) | |
| Sex | ||
| Male/female | 104/46 | |
| Timing of tumor fever | ||
| Before cancer detected | 24 | 16 |
| After cancer detected | 126 | 84 |
| Primary tumor | ||
| Genitourinary | 44 | 29 |
| Gastrointestinal | 41 | 27 |
| Lung | 22 | 15 |
| Head and neck | 19 | 13 |
| Others | 24 | 16 |
| Disease extent | ||
| Metastatic disease | 102 | 68 |
| Locoregionally advanced disease | 48 | 32 |
| Performance status | ||
| 0–1 | 78 | 52 |
| ≥2 | 72 | 48 |
| Initial presentation | ||
| Fever found just after admission | 90 | 60 |
| Suspected infection before admission | 37 | 25 |
| Fever found after cessation of dexamethasone | 21 | 14 |
| Fever of unknown origin before admission | 2 | 1 |
| Antibiotics use | ||
| First line | 119 | 79 |
| Second or third line | 31 | 21 |
| Febrile episode to NF suspicion | ||
| ≤7 days | 93 | 62 |
| 8–14 days | 44 | 29 |
| >14 days | 13 | 9 |
Ninety (60%) patients were unaware of any fevers until it was recorded in hospital. Thirty-six (25%) were admitted to the hospital because of suspected infection. Twenty-one (14%) developed fever after dexamethasone cessation, including 19 after chemotherapy and two after radiotherapy. Two (1%) patients exhibited fever of unknown origin. Most (79%) patients received a broad-spectrum (first line) antibiotic combination, including cephalothin, gentamicin, clindamycin, and penicillin, during their febrile periods. The interval between first febrile episode and NF suspicion ranged from five days to five weeks; 93 (62%) were suspected of having NF within seven days.
The characteristics of NF are listed in Table 2. The highest peak temperature ranged from 38.0 to 40.8°C (39.0±0.6°C). The most common daily peak temperatures were between 38.0 and 38.9°C. Twenty-five (17%) patients experienced chills. One hundred thirty-nine (93%) patients showed no increase in baseline pulse rates except during febrile periods; the fluctuations mainly followed body temperature alterations (Fig. 1, Fig. 2). The remaining 11 (7%) patients had transiently elevated baseline pulse rates at the time of cessation of postchemotherapy dexamethasone (Fig. 3a). The white blood counts ranged from 2,700 to 36,700/μL; 66 (44%) patients had normal values. Twelve (8%) patients had band forms of more than 5%, of which six represented leukoerythroblastosis. Other paraneoplastic syndromes were noted in 38 (25%) cases, including 14 with leukemic-like reactions, 13 with hypercalcemia, two with hypercalcemia plus leukemic-like reactions, seven with disseminated intravascular coagulation, one with dermatomyositis, and one with cachexia.
Table 2. Characteristics of NF According to Clinical and Laboratory Findings
| Characteristics | Cases | % |
|---|---|---|
| Peak body temperature | ||
| Range | 38.0–40.8°C | |
| Median | 39.0°C | |
| Mean | 39.0°C | |
| Usual daily peak temperature | ||
| 37.0–37.9°C | 43 | 29 |
| 38.0–38.9°C | 103 | 69 |
| 39.0–39.9°C | 4 | 3 |
| Associated with chillness | ||
| Yes | 25 | 17 |
| No | 125 | 83 |
| Leukocyte counts (per μL) | ||
| <4,000 | 6 | 4 |
| 4,000–9,990 | 66 | 44 |
| 10,000–19,990 | 60 | 40 |
| 20,000–29,990 | 16 | 11 |
| >30,000 | 2 | 1 |
| Band form (n=142) | ||
| <5% | 134 | 89 |
| ≥5% | 12 | 8 |
| Not check | 4 | 3 |
| Baseline pulse rate elevation during febrile periods | ||
| Limited to febrile time | 139 | 93 |
| Not limited to febrile time | 11a | 7 |
| Associated with other paraneoplastic syndrome | ||
| Yes | 38 | 25 |
| No | 112 | 75 |
aOccurring after cessation of postchemotherapy dexamethasone. |
Fig. 1
NF pattern. a) Once-daily single fever spikes, with peak time mainly at 9 am. b) Once-daily fever spike mingling with a few twice-daily fever spikes, with peak time mainly at 5 pm.
Fig. 2
NF pattern. a) Twice-daily fever spikes, with peak time mainly at 9 am or 9 pm. b) Twice-daily fever spikes, with peak time mainly at 9 am or 9 pm.
Fig. 3
Comparison of NF and infectious fever. a) NF: fever patterns showed transient pulse rates elevation at the time of postchemotherapy (CT) dexamethasone cessation. b) Infectious fever: fever patterns showed significant pulse rates elevation at the time of infectious period.
The patterns of NF are listed in Table 3. All patients exhibited intermittent fever patterns. Once-daily fever spike patterns occurred in 108 (72%) patients, of which 83 presented pure once-daily fever spikes (Fig. 1a) and 25 had once-daily fever spikes mingling with a few twice-daily fever spikes (Fig. 1b). Fever spikes were most commonly found at 9 am (42%) and 5 pm (37%). Twice-daily fever spike patterns (Fig. 2) were noted in the 42 (28%) remaining patients; 9 am and 9 pm (36%) and 9 am and 5 pm (31%) were the most common.
Table 3. Patterns of NF
| Cases | % | |
|---|---|---|
| Intermittent fever spike | 150 | 100 |
| Once-daily fever spike patterns | 108 | 72 |
| Fever spike, mainly | ||
| 1 am | 1 | 1 |
| 5 am | 3 | 3 |
| 9 am | 45 | 42 |
| 1 pm | 3 | 3 |
| 5 pm | 40 | 37 |
| 9 pm | 5 | 5 |
| Uncertain | 11 | 10 |
| Twice-daily fever spike patterns | 42 | 28 |
| Fever spike, mainly | ||
| 1 am and 1 pm | 1 | 2 |
| 1 am and 5 pm | 2 | 5 |
| 5 am and 5 pm | 6 | 14 |
| 5 am and 9 pm | 3 | 7 |
| 9 am and 5 pm | 13 | 31 |
| 9 am and 9 pm | 15 | 36 |
| Uncertain | 2 | 5 |
Fever persisted even after adequate antibiotic treatment. When NF was suspected, patients initially received naproxen 125mg every 12 hours. For 23 patients whose fever did not resolve, the naproxen dose was increased to 250mg every 12 hours because of no complete lysis of fever. After the naproxen dosage adjustment, patients had a complete and sustained defervescence within 24 hours (Fig. 4a) in 130 (87%), a partial lysis in 15 (10%), and a failure to defervesce in 5 (3%). One hundred eleven patients, including 17 who did not respond well to naproxen, received chemotherapy later. No fever developed after the administration of chemotherapy, but all chemotherapy included dexamethasone for emesis prevention. Follow-up periods ranged from two weeks to 43 months. Of these, with the exception of 24 patients who were lost to follow-up, 122 died. The median survival time using Kaplan-Meier methods was 12 weeks.
Fig. 4
Comparison of NF and infectious fever. a) NF: complete fever lysis after naproxen given. b) Infectious fever: complete fever lysis after control of infection.
We also collected data on 60 oncology patients with septicemia as a comparison group. Among them, 43 (72%) showed gram-negative bacilli bacteremia. Fifteen (25%) had polymicrobial septicemia. The highest peak temperature ranged from 37.2 to 40.5°C (38.7±0.9°C). Eleven (18%) patients experienced chills. Five (8%) patients had hypotension. Thirty (50%) patients had symptoms of consciousness disturbance and/or dyspnea. The white blood counts ranged from less than 200 to 87,300/μL; eight (13%) had white blood cell counts of less than 1,000/μL. Sixteen (27%) patients had band forms of more than 5%.
Comparison of fever patterns between patients with NF and those with infections, from vital sign flow sheets, is listed in Table 4. Twenty (33%) patients had intermittent fever patterns recorded. Intermittent fever patterns were seen more frequently in patients with NF than in patients with infections (P<0.001). In contrast to patients with NF, all patients with infections showed an increase in baseline pulse rates during febrile periods (Fig. 3b) (P<0.001).
Table 4. Comparison of Fever Pattern Between NF and Infection From Vital Sign Flow Sheets
| NF (n=150) | Infection (n=60) | P-value | |||
|---|---|---|---|---|---|
| Cases | % | Cases | % | ||
| Peak body temperature | 38.0–41.6°C | 37.2–40.5°C | |||
| Median | 39.0°C | 38.6°C | |||
| Mean | 39.0°C | 38.7°C | |||
| Baseline pulse rate elevation during febrile period | <0.001 | ||||
| Limited to febrile time | 139 | 93 | 0 | 0 | |
| Not limited to febrile time | 11 | 7 | 60 | 100 | |
| Intermittent fever pattern | <0.001 | ||||
| Yes | 150 | 100 | 20 | 33 | |
| No | 0 | 0 | 40 | 67 | |
Seventeen (28%) patients in this group had current naproxen use. Their highest peak temperature was 38.7±1.0°C (37.4–41.0°C). Forty-three (72%) did not use naproxen; highest peak temperature was 38.7±0.8°C (37.2–41.0°C). This difference was not statistically significant (P=0.75). Twenty-nine (48%) had their infections controlled; their pulse rates all returned to their baseline (Fig. 4b).
Discussion
The mechanism of NF production is related to cytokines. Cancer cells can spontaneously produce cytokines, and some cytokines released by neoplastic cells are pyrogenic. These pyrogenic cytokines include tumor necrosis factor-alpha (TNF-α), interleukin-1 (IL-1), IL-6, and interferon. The cytokines promote prostaglandin E2 (PGE2) formation that acts on the hypothalamic thermoregulatory center causing an elevation of temperature set point.2, 13, 20
NF occurs more commonly in metastatic or advanced stages of malignant disease, and in a few patients, it also is associated with other paraneoplastic syndromes.4, 21 Approximately 60% of patients did not detect fever until hospitalization, but a high peak fever, chills, and leukocytosis could not exclude the diagnosis of NF. Before a diagnosis of NF can be made, antibiotic administration, vital signs, and blood and other cultures monitoring are essential to exclude possible fatal infections.
All study patients with NF exhibited intermittent fever patterns. Once-daily fever spike patterns occurred in 72% of our patients. Fever spikes could occur at any time, but 9 am and 5 pm were the most common. Twice-daily fever spike patterns were noted in 28% of the remaining patients; 9 am and 9 pm and 9 am and 5 pm were the most common. The exact mechanism of this NF timing remains unclear. Circadian variation of body temperature, both normal and febrile, is a well-known fact. Temperature curves showed statistically significant circadian rhythms, with higher values at night and lower ones during early morning.22 Serum cortisol is known to peak during morning and maintain low levels at night.22 The presence of circadian rhythms of cytokines has been proven in people with advanced neoplasms.23 The levels of cortisol and TNF-α displayed statistically significant circadian fluctuations in gastrointestinal cancer.23, 24 The presence of complex self-regulation mechanisms among the neuroendocrine system, the hypothalamic-pituitary-adrenal axis, and the cytokine network in advanced cancer patients also has been suggested.25, 26, 27 The elevation of serum cytokines correlated with cortisol rhythm has been demonstrated in patients with metastatic colorectal cancer.25 Advanced neoplastic disease with elevation of serum cytokines may be associated with enhanced cortisol levels and alterations of its circadian secretion.26
Cancer and its treatment can lead to increase in the release of proinflammatory cytokines, flatten the slope of cortisol, decrease the sensitivity of glucocorticoid, decrease cortisol responsiveness to stress, and activate the cytokine signal pathway, including nuclear factor-kappa B (NFκB) and p38 mitogen-activated protein kinase in the brain.27 The circadian fluctuations of cytokine, decrease in cortisol sensitivity, and their cytokine signal pathway activation in the brain are all probably contributory factors that influence NF patterns and their daily peak temperatures.
Using intermittent fever alone to diagnose NF is inadequate, as fever may be because of infection.16, 17 Monitoring pulse rate change is mandatory. At the onset of infection, bacteria rapidly replicate and release toxins to stimulate the immune system, leading to cytokine release.28 When systemic inflammatory response syndrome occurs, clinical symptoms can include fever, tachycardia, and leukocytosis.29 One-third of our cancer patients with septicemia revealed intermittent fever patterns, but they all had a rise in their baseline pulse rate during febrile periods.
Unlike bacterial infections, most of our cases did not show increases in their baseline pulse rates except during febrile periods. Pulse rate change mainly followed body temperature fluctuations. The presence of the circadian rhythm of cytokine and decrease in cortisol sensitivity are probably the explanation. Tumor progression, cancer therapy, and cancer complications increase the expression of inflammatory cytokines and NF-κB activation.27, 30, 31, 32 Baseline pulse rate is not definitive without several days of monitoring. Cancer chemotherapy can increase the production of cytokines.30 Cessation of dexamethasone does not suppress cytokines.10, 23 Seven percent of our patients had transiently elevated pulse rates at the time of discontinuation of postchemotherapy dexamethasone. In these circumstances, fever origin from infection or neoplasia is difficult to differentiate (Fig. 3).
Both naproxen and corticosteroids have an antipyretic effect on NF.2, 10 In malignancy, the presence of NFκB actually increases cytokine products. NSAIDs are effective in ameliorating fever by inhibiting cyclooxygenase activity and suppressing NFκB activation.8, 9, 10, 11, 12, 13, 14, 15, 31, 33 Naproxen-induced complete lysis of fever in 87% of our patients matches the results from studies by Chang and Economos.8, 9, 10, 11 Corticosteroids may block the febrile response through dual mechanisms, including reduction of PGE2 synthesis by inhibiting the activity of phospholipase A2 and the blockade of the transcription of the mRNA for pyrogenic cytokine.10, 23 Dexamethasone did mask NF in 14% of our patients receiving radiotherapy and chemotherapy. Dexamethasone is also a common antiemetic agent for cancer chemotherapy. Our patients, including those who did not respond well to naproxen, showed complete lysis of fever after the administration of chemotherapy. Data from the comparison group revealed that naproxen use has no influence on septic fever occurrence.
Ten percent to 20% of patients with NF failed to exhibit a complete lysis of fever after the administration of naproxen.8, 11 An NSAID trial yielded complete fever lysis in 38% of patients without neoplastic disease.8, 14 We hope that vital sign flow sheets, showing fever patterns together with pulse rate variation curves, can help to provide clinicians with clues to find the fever etiologies in these patients.
In conclusion, NF pattern is characterized by intermittent fever without an obvious increase in baseline pulse rate except during febrile periods. The vital sign flow sheet is a useful tool to help clarify early suspicion of NF in oncology patients. The recognition of NF patterns from vital sign flow sheets can help to identify NF and differentiate between other causes of fever in cancer patients. The limitation of vital sign flow sheets is that baseline pulse rate cannot be known without several days' monitoring. Transient pulse rate elevation could occur at the time of cessation of postchemotherapy dexamethasone. Studies that correlate serum cytokine change and NF pattern are needed.
Acknowledgments
The authors thank their oncology nursing staff for their help in recording patient data on the vital sign flow sheets.
References
- . Neoplastic fever: all who shiver are not infected. Support Care Cancer. 2005;13:863–864
- . Neoplastic fever: a neglected paraneoplastic syndrome. Support Care Cancer. 2005;13:870–877
- . Infections or neoplasm as causes of prolonged fever in cancer patients. Am J Med Sci. 1976;272:65–74
- Are there criteria for the diagnosis of paraneoplastic fever?. Rev Med Interne. 2000;21:684–692
- . Fever and solid tumor: diagnostic value of procalcitonin and C-reactive protein. Rev Med Interne. 2001;22:706–714
- C-reactive protein and erythrocyte sedimentation rate in differential diagnosis between infections and neoplastic fever in patients with solid tumours and lymphomas. Support Care Cancer. 2001;9:124–128
- Causes of fever and value of C-reactive protein and procalcitonin in differentiating infections from paraneoplastic fever. Support Care Cancer. 2004;2:593–598
- . Neoplastic fever responds to the treatment of an adequate dose of naproxen. J Clin Oncol. 1985;3:552–558
- . Utility of naproxen in the differential diagnosis of fever of undetermined origin in patients with cancer. Am J Med. 1984;76:597–603
- . Antipyretic effect of naproxen and corticosteroids on neoplastic fever. J Pain Symptom Manage. 1988;3:141–144
- . The effect of naproxen on fever in patients with advanced gynecologic malignancies. Gynecol Oncol. 1995;56:250–254
- A randomized trial of the effect of three non-steroid anti-inflammatory agents in ameliorating cancer-induced fever. J Intern Med. 1990;228:451–455
- . Neoplastic fever. Palliat Med. 1996;10:217–224
- . Lack of value of the naproxen test in the differential diagnosis of prolonged febrile illnesses. Am J Med. 2003;115:572–575
- . Intermittent fever of neoplastic origin. Rev Prat. 2002;52:145–148
- . Intermittent fever of infectious origin. Rev Prat. 2002;52:139–144
- . Symptomatic intermittent fever of inflammatory diseases. Rev Prat. 2002;52:160–166
- . A retrospective review of 226 hospitalized patients with fever. Intern Med. 2007;46:17–22
- . Drug-induced fever. Drug Intell Clin Pharm. 1986;20:413–420
- . Fever. Semin Oncol. 1997;24:288–298
- . Tumor fever in patients with nasopharyngeal carcinoma: clinical experience of 67 patients. Am J Clin Oncol. 1998;21:422–425
- Circadian variation of human acute phase response. Arch Med Res. 1996;27:157–163
- . The estimation of endogenous tumor necrosis factor alpha and cortisol levels in serum in advanced neoplasm. J Exp Clin Cancer Res. 1999;18:241–245
- . Dynamics of circadian fluctuations in serum concentration of cortisol and TNF-alpha soluble receptors in gastrointestinal cancer patients. Oncol Rep. 2001;8:207–212
- Elevated serum cytokines correlated with altered behavior, serum cortisol rhythm, and dampened 24-hour rest-activity patterns in patients with metastatic colorectal cancer. Clin Cancer Res. 2005;11:1757–1764
- Immune and endocrine mechanisms of advanced cancer-related hypercortisolemia. In Vivo. 2007;21:647–650
- Neuroendocrine-immune mechanisms of behavioral comorbidities in patients with cancer. J Clin Oncol. 2008;26:971–982
- Bacterial invasion augments epithelial cytokine responses to Escherichia coli through a lipopolysaccharide-dependent mechanism. J Immunol. 2001;166:1148–1155
- . Sepsis and septic shock. In: Braunwald E, Fauci AS, Kasper DL, et al. editor. Harrison's principles of internal medicine. 15th ed.. New York: McGraw Hill; 2001;p. 799–804
- The host environment promotes the constitutive activation of nuclear factor-kappaB and proinflammatory cytokine expression during metastatic tumor progression of murine squamous cell carcinoma. Cancer Res. 1999;59:3495–3504
- . Cancer chemotherapy-related symptoms: evidence to suggest a role for proinflammatory cytokines. Oncol Nurs Forum. 2006;33:535–542
- . Is NF-kappaB a good target for cancer therapy? Hopes and pitfalls. Nat Rev Drug Discov. 2009;8:33–40
- . Nonsteroidal anti-inflammatory agents differ in their ability to suppress NF-kappaB activation, inhibition of expression of cyclooxygenase-2 and cyclin D1, and abrogation of tumor cell proliferation. Oncogene. 2004;23:9247–9258
PII: S0885-3924(10)00328-3
doi:10.1016/j.jpainsymman.2010.01.015
Crown Copyright © 2010. Published by Elsevier Inc. All rights reserved.
Volume 40, Issue 2 , Pages 256-265, August 2010
