A Dukes/Jass combination--is it more discriminating?

BACKGROUND: In 1987 Jass described a modified staging system for colorectal cancer using two anatomical criteria in common with the Dukes system; extent of spread through the bowel wall and the presence or absence of lymph node involvement. Additionally it used two 'biological' criteria; the nature of the expanding margin of the tumour (pushing or infiltrating), and the presence or absence of lymphocytic infiltration within the tumour. This study aims to determine whether a combination of the Dukes and Jass staging systems provides a better predictor of five year survival in patients with colorectal cancer than Dukes stage alone. METHOD: The Dukes and Jass stages along with vital status at five years were recorded for all 612 patients undergoing resection for colorectal cancer at the Royal Bolton Hospital and the Beaumont (BMI) Hospital, Bolton between 1991 and 1998. Kaplan-Meier survival curves with log rank test were used to show how survival correlated with Jass group stratified by Dukes stage. RESULTS: Both the Dukes B and the Dukes C tumours could be divided into groups with significantly different five year survival rates when stratified by Jass group. Five year survival for Dukes stage B, Jass group II tumours was 73.74% compared to Dukes B Jass III tumours whose survival was 51.38% (P = 0.0018). Five year survival for Dukes C Jass III tumours was 43.18% and for Dukes C Jass IV survival was 24.39% (P = 0.0029). CONCLUSION: By combining the biological criteria of the Jass staging system with the anatomically based Dukes system, both Dukes B and C tumours can be divided into groups with significantly different five year survival figures.

Absence of socioeconomic variation in survival from colorectal cancer in patients receiving surgical treatment in one health district: cohort study.

OBJECTIVE: To examine whether there is an association between patient deprivation status and survival from colorectal cancer among patients receiving treatment of the same type and quality. PATIENTS AND METHODS: A survival study was conducted of all colorectal cancer patients diagnosed between 1991 and 1997 who received surgery either in the NHS district general hospital or the private hospital of one UK health district. The five-year survival rates, both all cause and colorectal cancer specific, were calculated for subgroups defined by patient age, gender, stage and deprivation status using Kaplan-Meier curves. Cox proportional hazards models were used to examine the influence of deprivation on five-year survival after adjusting for age, gender and stage. RESULTS: There were 603 consecutive colorectal patients during the study period. Five-year all-cause and colorectal cancer-specific survival rates were 41% and 53%, respectively. There was no association between deprivation status and stage at diagnosis (P = 0.308). Multivariable proportional hazards modelling (adjusting for gender, age and tumour stage) demonstrated no association between deprivation status and survival. CONCLUSION: In this single district study, no relationship between patient socioeconomic status and survival from colorectal cancer could be demonstrated. Consistency in the type and quality of treatment offered to patients by the same clinical teams may have been responsible for the equitable survival outcomes.

Anal function following third degree tears.

OBJECTIVE: To assess the anal function of women who have suffered a third degree perineal tear during parturition. PATIENTS: Fifty-three consecutive women who sustained a third degree tear, between January 1998 and March 2000, at the Princess Anne Maternity Unit, Royal Bolton Hospital were assessed. METHODS: Women were assessed at 3 months post partum using the Cleveland Clinic Incontinence Score, digital assessment of anal sphincter and endo-anal ultrasound scan. RESULTS: At 3 months post partum 75% of the participants had no symptoms of anal incontinence, 18% had mild symptoms and 7% had more severe symptoms. Anal endosonography demonstrated normal anal sphincters in 66% of participants, an abnormality in the external sphincter in 29% and a defect in both sphincters in 2%. There was poor correlation between symptoms and scan defects. CONCLUSION: The incidence of anal incontinence following repair of a third degree tear was not high and it is unlikely that we are missing a hidden pool of symptomatic women. No major change in management policy is required. The routine assessment of anal function in women who had sustained a third degree tear was appreciated by the women and enabled us to identify the small portion of women with significant symptoms.

The value of animal models in the development of new drugs for the treatment of the sepsis syndrome.

The sepsis syndrome has proved to be one of the most difficult clinical situations to replicate in animal models. The outcome of sepsis in patients depends not only on the pathogens, sizes of bacterial challenge and adequacy of treatment, but also on age, host defence mechanisms and individual response to septic shock and multiple organ failure. Novel immunomodulatory agents may control endotoxaemia or bacteraemia in animal models but they have not proved effective in septic patients. This article considers whether the principal problem lies in our interpretation of the data from animal studies, or whether such models are unsuitable for assessing new treatments for the sepsis syndrome.

Metabolism of sepsis and multiple organ failure.

"Septic autocannabalism" been coined to describe the metabolic response that follows severe sepsis in humans. The normal protein- and energy-conserving mechanisms evoked during simple starvation are not observed following the onset of sepsis. The metabolic response to sepsis entails rapid breakdown of the body's reserves of protein, carbohydrate, and fat. Hyperglycemia with insulin resistance, profound negative nitrogen balance, and diversion of protein from skeletal muscle to splanchnic tissues are prominent features. These responses are believed to be mediated in large part by inflammatory cytokines such as tumor necrosis factor alpha (TNFalpha), interleukin 1beta (IL-1beta), and IL-6. Secondary induction of catecholamines, cortisol, and glucagon by cytokines is likely to be another important effector mechanism. Infection and inflammation elicit a complex network of interwoven responses, and no single mediator alone accounts for the responses observed. Sepsis also commonly involves alterations in cardiovascular function with altered flow to key metabolic sites, hypoxia, damage to the gut's mucosal barrier, secondary organ failure, and alterations in capillary permeability. These structural and functional alterations also strongly influence the metabolic profile during infection. If these catabolic responses persist for more than a few days, severe malnutrition results and is likely to be an important risk factor for mortality in these patients. The altered metabolic milieu during sepsis prevents effective use of exogeneously delivered glucose and protein; at best, administration of these agents ameliorates but does not prevent the persistence of catabolism. Delivery of agents that antagonize cytokines and other moieties such as glutamine and growth hormone may, in the future, help to restore nitrogen balance during sepsis.

Effects of in vivo endotoxin infusions on in vitro cellular immune responses in humans.

Studies of the immune response of patients following major injury have identified significant abnormalities, some of which may be due to the effects of endotoxin. To evaluate the effect of endotoxin on the immune system without conflicting variables, we studied 18 normal, healthy male volunteers each on two occasions. In one study, Escherichia coli endotoxin was administered intravenously at a dose of 4 ng/kg. In the other, saline was given. Blood for immune function studies was obtained at either 0, 4, or 24 hr (seven volunteers), 0, 1, and 4 hr (five volunteers), or 0, 4, and 6 hr (six volunteers) postinfusion. Peripheral blood mononuclear cells (PBMC) were isolated and adjusted to the same concentration. Measurements following endotoxin infusion were compared with those of the same volunteers following saline infusion and with those from normal ambulatory laboratory volunteers. Interleukin 1 (IL-1) production by adherent cells was significantly reduced at 1 hr post endotoxin infusion. Significant decreases in number of mononuclear cells, response to phytohemagglutinin (PHA), and production of IL-2 and IL-1 were observed by 4 hr after endotoxin infusion. No significant changes in percentages of monocytes, lymphocytes, or CD3, CD4, or CD8 lymphocytes were observed at any time. By 24 hr postinfusion all values had returned to normal or, in some cases, supranormal levels. Response to PHA by PBMC from volunteers 4 hr following endotoxin was completely restored by in vitro addition of recombinant human IL-2 but was only marginally improved by IL-1. In vitro addition of indomethacin to PBMC cultures responding to PHA reduced the suppression observed after in vivo endotoxin but also was not as effective as IL-2. In a fourth study, seven volunteers were treated as above either with two doses (800 mg each) of the cyclooxygenase inhibitor ibuprofen before endotoxin infusion or with ibuprofen alone. Ibuprofen pretreatment completely restored the PBMC response to PHA to normal and caused a significant decrease in the endotoxin-induced suppression of IL-2 production. However, the decrease in circulating PBMC number and adherent cell secretion of IL-1 was not affected by inhibition of the cyclooxygenase pathway. These results suggest that endotoxin has immunomodulatory effects on both adherent mononuclear-cell and T-lymphocyte function and that more than one mechanism is involved.

Both cyclooxygenase-dependent and cyclooxygenase-independent pathways mediate the neuroendocrine response in humans.

The neuroendocrine response that occurs after operation, trauma, or infection is essential for maintaining homeostasis within the host. Corticotropin-releasing hormone, (CRH), arginine vasopressin (AVP), and products of the cyclooxygenase pathway have been proposed as possible mediators of the pituitary-adrenal response. We studied the relationship between the induction of these mediators and the onset of the neuroendocrine response in healthy male volunteers receiving one of two standard provocative stimuli: (1) Escherichia coli endotoxin (4 ng/kg intravenously) and (2) hypoglycemia induced by insulin (0.15 units/kg intravenously). Additional subjects receiving these stimuli were pretreated with the cyclooxygenase inhibitor ibuprofen, 1600 mg orally. Serial measurements were made of circulating concentrations of CRH, AVP, adrenocorticotropic hormone, cortisol, and catecholamines. A sudden rise in circulating AVP (but not CRH) concentrations preceded the onset of all other responses after both stimuli. The prevention of this surge of AVP after endotoxin by ibuprofen was associated with blockade of the neuroendocrine response. These data demonstrate that two independent pathways are available for stimulation of the stress response. After both stimuli, peripheral AVP (but not CRH) levels may mirror alterations that occur within the central nervous system.

Sepsis, signals, and surgical sequelae (a hypothesis).

Cytokines, products of the bodies on cells, are the major signals that orchestrate the host's response to bacterial infection. Those signals generated following infection include interleukins 1, 2, and 6, interferon gamma, and tumor necrosis factor (or cachectin). Tumor necrosis factor is the only cytokine, to date, that has been shown to fulfill Koch's postulates and, thus, be casually related to host responses. Host responses to cytokines vary because of alterations in the genetic mechanism that controls cytokine production, because of an alteration in the responsiveness of the reticuloendothelial system at the time of signal induction and because of alterations in cell surface receptors. Only now are techniques evolving that can detect cytokine concentrations or production rates to relate these molecules to varying aspects of human disease. A major therapeutic goal in the future will be directed toward blocking the deleterious effects of cytokines while maintaining their protective or beneficial effects.

Circulating interleukin-1 and tumor necrosis factor in septic shock and experimental endotoxin fever.

Interleukins (IL) -1 beta and -1 alpha and tumor necrosis factor (TNF-alpha) were measured by radioimmunoassay in plasma samples from 44 healthy individuals, 15 patients in septic shock, and 6 volunteers infused with endotoxin. Plasma IL-1 alpha levels were low (40 pg/ml) or undetectable in all situations. In 67% of the healthy subjects, plasma IL-1 beta levels were less than 70 pg/ml. Septic patients had higher plasma IL-1 beta levels (120 +/- 17 pg/ml, P = .001); those of surviving patients were higher than those of patients who died (P = .05). Plasma TNF-alpha concentrations in septic individuals were elevated (119 +/- 30 pg/ml) and correlated with severity of illness (r = .73, P = .003), but no correlation was observed between plasma IL-1 beta and TNF-alpha concentrations in individual samples. Infusion of endotoxin caused a twofold elevation of IL-1 beta, from a baseline of 35 +/- 5 pg/ml to a maximum of 69 +/- 27 pg/ml at 180 min (P less than .05). Peak TNF-alpha levels after endotoxin infusion were 15 times higher than IL-1 beta levels, were attained more rapidly (90 min), and as with the septic patients, did not correlate with IL-1 beta levels. These data support the concept that plasma IL-1 beta and TNF-alpha concentrations are regulated independently and are associated with different clinical outcomes.

Tumour necrosis factor and bacterial sepsis.

This review examines the evidence that tumour necrosis factor is the principal mediator of the derangements associated with severe sepsis.

Ibuprofen causes reduced toxic effects of interleukin 2 administration in patients with metastatic cancer.

Metastatic cancer was treated with interleukin 2 and lymphokine-activated killer cells with the addition of the cyclooxygenase inhibitor ibuprofen in an attempt to reduce side effects in 13 patients (eight male and five female). Twenty-six patients treated with only interleukin 2 and lymphokine-activated killer cells formed the control group. After interleukin 2 administration, a significantly increased number of lymphokine-activated killer cells were transfused in ibuprofen-treated patients. Cytotoxic effects were not significantly different in the treated and untreated groups. With regard to cell phenotype, both groups of patients manifested significant activation of the immune system as measured by T10 and OK1a. Symptom scores were dramatically reduced in patients treated with ibuprofen. Temperature above 37 degrees C were rare. Ibuprofen did not significantly alter rate of response in this immunotherapy trial (38% vs 42%). Ibuprofen is now routinely used in all of our current immunotherapy trials.

Chronic TNF infusion causes anorexia but not accelerated nitrogen loss.

It has been proposed that many of the physiologic and metabolic changes that occur during critical illness and malignancy are mediated by the cytokine tumor necrosis factor alpha/cachectin (TNF). To test this hypothesis, a study of the metabolic responses that occurred during 5 days of continuous intravenous (I.V.) infusion of TNF both in rats and tumor-bearing humans was conducted. TNF administration was associated with anorexia, fluid retention, acute phase responses, and negative nitrogen balance. In both species, changes in nitrogen balance were related to the onset of anorexia and not to the development of hypermetabolism and accelerated net tissue breakdown. TNF may represent the primary afferent stimulus inducing many of the metabolic changes that occur during critical illness, but it is not solely responsible for the accelerated net proteolysis that occurs in these patients.

A single dose of endotoxin increases intestinal permeability in healthy humans.

To investigate the effects of endotoxin on gut barrier function, we performed paired studies of intestinal permeability in healthy humans (N = 12) receiving intravenous Escherichia coli endotoxin (4 ng/kg) or 0.9% saline solution. Two nonmetabolizable sugars, lactulose and mannitol, which are standard permeability markers, were administered orally, 30 minutes before and 120 minutes after the test injection. The 12-hour urinary excretion of these substances after endotoxin/saline solution administration was used to quantitate intestinal permeability. After endotoxin administration systemic absorption and excretion of lactulose increased almost two-fold (mean +/- SEM, 263 +/- 36 mumol per 12 hours vs 145 +/- 19 mumol per 12 hours during saline studies). Similar but less marked alterations in mannitol absorption and excretion occurred after endotoxin injection (5.7 +/- 0.3 mmol per 12 hours vs 4.9 +/- 0.3 mmol per 12 hours). When individual 12-hour lactulose excretion after endotoxin administration was related to the magnitude of systemic responses, a significant relationship occurred between lactulose excretion and elaboration of norepinephrine and between lactulose excretion and minimum white blood cell count. These data suggest that a brief exposure to circulating endotoxin increases the permeability of the normal gut. These observations are consistent with the hypothesis that during critical illness, prolonged or repeated exposure to systemic endotoxins or associated cytokines may significantly compromise the integrity of the gastrointestinal mucosal barrier.

Interleukin-2 initiates metabolic responses associated with critical illness in humans.

The cytokine interleukin-2 is a primary modulator of the immune response that occurs after infection, trauma, and transplant rejection, yet its role as a mediator of associated metabolic changes in surgical illness is unknown. We studied clinical and metabolic responses in eleven tumor-bearing humans with normal renal and hepatic function receiving bolus intravenous (I.V.) interleukin-2 (30,000 U/kg). Additional subjects (n = 6) were pretreated with the cyclooxygenase inhibitor, ibuprofen (1600 mg, orally), before interleukin-2 administration. Serial measurements were made of vital signs, symptoms, hematology, and plasma concentrations of pituitary and stress hormones and selected cytokines. Administration of interleukin-2 resulted in fever, tachyacardia, "flu-like" symptoms, and neurohormonal elaboration. The responses observed were quantitatively similar to those that occurred after endotoxin administration in healthy subjects (n = 13), but differed in the following manner: 1) the onset of fever and endocrine changes occurred after a longer latent interval (180-240 minutes vs. 60-90 minutes after endotoxin), 2) peak responses after the administration of interleukin-2 also occurred later, 3) no increased circulating tumor necrosis factor was detected after administration of interleukin-2 (peak plasma concentration was greater than 35 pg/ml vs. 270 +/- 70 pg/ml after endotoxin administration), and 4) administration of interleukin-2 but not of endotoxin was associated with increased circulating concentrations of gamma interferon (peak plasma concentration 1.7 +/- 0.2 NIH U/ml vs. less than 0.1 NIH U/ml after endotoxin administration). Fever and neurohormonal responses after interleukin-2 administration were greatly attenuated by ibuprofen administration. Interleukin-2 induces other cytokines that exert their effects largely through the cyclooxygenase pathway. Interleukin-2 may be an important signal, initiating the integrated host responses to infection and injury.

Tumor necrosis factor and endotoxin induce similar metabolic responses in human beings.

After injury, infection, or major operations a number of predictable metabolic responses occur. It has been proposed that the cytokine tumor necrosis factor (TNF)/cachectin is a primary mediator of these host responses. To test this hypothesis, we studied 16 tumor-bearing humans with normal renal and hepatic function, who received 24-hour continuous intravenous infusions of escalating doses of recombinant TNF (4 to 636/micrograms/m2/24 h). Serial measurements were made of vital signs and plasma concentrations of TNF, interleukin-1, adrenocorticotropic hormone, cortisol, iron, glucose, and C-reactive protein. Low doses of TNF had minimal metabolic effects, but infusions of greater than or equal to 545 micrograms/m2/24 hr (n = 8) resulted in fever, pituitary, and stress hormone release and acute phase changes. These alterations were compared with the changes that occurred in healthy humans (n = 13) receiving intravenous bolus injections of Escherichia coli endotoxin (4 ng/kg). TNF infusion in doses greater than or equal to 545 micrograms/m2/24 hr produced peak plasma TNF concentrations and metabolic responses that were similar to those after endotoxin injection. Interleukin-1 concentrations remained basal after TNF or endotoxin administration. TNF may represent the primary afferent signal that initiates many of the metabolic responses associated with sepsis and endotoxemia.

Detection of circulating tumor necrosis factor after endotoxin administration.

Cytokines, products of stimulated macrophages, are thought to mediate many host responses to bacterial infection, but increased circulating cytokine concentrations have not been detected consistently in infected patients. We measured plasma concentrations of circulating tumor necrosis factor alpha (cachectin), interleukin-1 beta, and gamma interferon, together with physiologic and hormonal responses, in 13 healthy men after intravenous administration of Escherichia coli endotoxin (4 ng per kilogram of body weight) and during a control period of saline administration. Eight additional subjects received ibuprofen before receiving endotoxin or saline. Plasma levels of tumor necrosis factor were generally less than 35 pg per milliliter throughout the control period, but increased 90 to 180 minutes after endotoxin administration to mean peak concentrations of 240 +/- 70 pg per milliliter, as compared with 35 +/- 5 pg per milliliter after saline administration. Host responses were temporally associated with the increase in circulating tumor necrosis factor at 90 minutes, and the extent of symptoms, changes in white-cell count, and production of ACTH were temporally related to the peak concentration of tumor necrosis factor. Ibuprofen pretreatment did not prevent the rise in circulating tumor necrosis factor (mean peak plasma level, 170 +/- 70 pg per milliliter) but greatly attenuated the symptoms and other responses after endotoxin administration. Concentrations of circulating interleukin-1 beta and gamma interferon did not change after endotoxin administration. We conclude that the response to endotoxin is associated with a brief pulse of circulating tumor necrosis factor and that the resultant responses are effected through the cyclooxygenase pathway.

Inhibition of cyclo-oxygenase attenuates the metabolic response to endotoxin in humans.

Acute infection initiates fever, acute-phase changes, and catabolic responses in the host, resulting in weight loss, hypermetabolism, and accelerated proteolysis. To test the hypothesis that cyclo-oxygenase inhibition might attenuate these responses, we administered Escherichia coli endotoxin intravenously to seven normal volunteers and to seven additional subjects pretreated with a cyclo-oxygenase inhibitor (ibuprofen). Control studies were also performed following administration of saline and ibuprofen alone. Vital signs, metabolic rate, and concentrations of pituitary and stress hormones, as well as those of other substrates, were serially measured. Endotoxin administration produced a response similar to an acute illness, with flulike symptoms, fever, tachycardia, increased metabolic rate, and stimulation of stress hormone release. These changes were markedly attenuated by cyclo-oxygenase inhibition. The leukocytosis, hypoferremia, and elevation of the C-reactive protein level induced by endotoxin were unaffected by cyclo-oxygenase inhibition. These data indicate that activation of the cyclooxygenase pathway is necessary to produce many of the metabolic changes observed during critical illness.