Procoagulant activity in whole blood from patients with breast and colorectal cancer

Whole blood procoagulant activity was determined by measuring the endotoxin-induced shortening of the celite-activated whole blood recalcification time in patients with breast cancer (n = 29), colorectal cancer (n = 18), benign breast disease (n = 26), benign colorectal disease (n = 10), normal volunteers (n = 17) and surgical in-patients with non-malignant and non-inflammatory conditions (n = 18). Using this method, patients with breast and colorectal cancer produced significantly more procoagulant activity than normal controls (P less than 0.001), surgical in-patients (P less than 0.005) and patients with benign breast (P less than 0.001) and benign colorectal (P = 0.05) disease respectively. The difference between subjects classified as 'cancer' or 'non-cancer' was highly significant (P less than 0.0001). There were no significant differences in total white cell or absolute monocyte counts between subject groups, and in individual patients, there was no correlation between these parameters and the procoagulant activity. It is concluded that the activated whole blood recalcification time is a more reproducible way of measuring whole blood procoagulant activity than the original technique, and that using this method, patients with cancer show higher procoagulant activity than corresponding benign controls.     

Fibrinopeptide A in acute leukemia: relationship of activation of blood coagulation to disease activity

Plasma fibrinopeptide A (FPA) levels were determined in 20 unselected adult patients with acute nonlymphocytic and lymphocytic leukemia. The mean FPA level in patients with active disease (15.0 ng/ml) was significantly higher than during clinical remission (2.4 ng/ml, p less than 0.01). Elevated FPA levels were observed in patients with all morphological forms of acute leukemia. In the group of patients in clinical remission, 20/47 FPA values remained elevated beyond the normal range, suggesting that low-grade intravascular coagulation was present even when no leukemic cells were observed. Sequential studies revealed reduction of FPA levels to the normal range in five patients who entered clinical remission after chemotherapy and rapid elevation of the levels in eight patients who entered relapse after clinical remission. FPA levels rose significantly in five patients studied during induction chemotherapy. Thus, subclinical activation of blood coagulation, as defined by elevation of plasma FPA level, may occur commonly in acute leukemia. Plasma FPA generation may relate to leukemic disease activity.     

Heparin abolishes the chemotherapy-induced increase in plasma fibrinopeptide A levels

PURPOSE, PATIENTS, AND METHODS: Blood coagulation abnormalities are common in patients with cancer, particularly after treatment with chemotherapeutic agents. Chemotherapy has been associated with an increased incidence of thromboembolic events, and patients treated with chemotherapy often develop evidence of local phlebitis, which may lead to loss of venous access. We have utilized the radioimmunoassay for plasma fibrinopeptide A (FPA) and in vitro FPA generation to assess the rate of in vivo blood coagulation and the level of plasma thrombin activity in 16 cancer patients treated with chemotherapy. Eight patients were treated twice, once with chemotherapy alone and once with chemotherapy after an intravenous infusion of heparin (5,000 U). RESULTS: Our results confirm that FPA levels are elevated in most cancer patients. Following chemotherapy, FPA levels were further increased within 45 minutes (mean FPA = 5.2 ng/mL before chemotherapy versus 8.3 ng/mL after chemotherapy, p less than 0.01) and were accompanied by an increase in the FPA generation rate. Infusion of heparin prior to chemotherapy significantly lowered plasma FPA levels and abolished post-chemotherapy FPA generation. CONCLUSION: These data suggest that patients receiving chemotherapy express thrombin-like activity in plasma and, therefore, may be at risk for clinically significant intravascular activation of coagulation. Heparin diminished the laboratory evidence of this chemotherapy-related coagulopathy and may have a role in the prevention of thromboembolic disorders in some cancer patients undergoing cytotoxic therapy.     

Fibrinopeptide A and the phosphate content of fibrinogen in venous thromboembolism and disseminated intravascular coagulation

Concentrations of plasma fibrinopeptide A (FPA) were measured by radioimmunoassay in 50 patients with venous thromboembolism or disseminated intravascular coagulation or both. A consistent discrepancy was observed in values obtained with two anti-FPA antisera. Analysis of extracts from plasma of these patients by high-performance liquid chromatography (HPLC) revealed the presence of a phosphorylated and an unphosphorylated form of the A peptide. Differences in concentrations of FPA measured with the two antisera could be accounted for by their different reactivity with phosphorylated FPA (FPA-P). The differences were abolished by treatment with alkaline phosphatase. A good correlation was observed between the FPA-P content of free A- peptide material and of fibrinogen in plasma as determined by HPLC (r = .88, P less than .001, n = 11). In patients with elevated FPA levels, the mean FPA-P content of fibrinogen was significantly higher (P less than .002, n = 13) than in patients with normal FPA levels (n = 8) and in healthy controls (n = 14). Phosphorus in fibrinogen did not correlate with fibrinogen degradation products or fibrinogen levels and became normal on adequate anticoagulation. Therefore, blood-clotting activation may lead to a high phosphate content of fibrinogen and of free FPA in plasma.     

Malignant Colorectal Strictures in Crohn''s Disease

One hundred thirty-two of 980 patients (13.5%) with Crohn's disease (CD) involving the colon, admitted to The Mount Sinai Hospital between 1959 and 1985, developed 175 colonic strictures. Thirty-three patients developed more than one stricture. The frequency was twice as great in colitis (19%) as in ileocolitis (11%). Ten malignant strictures were identified in nine patients (three ileocolitis, six colitis). One of these patients had three strictures (two malignant, and one benign), and two had two strictures (one malignant and one benign). The frequency of cancer in patients with stricture (6.8%) was higher than in those without stricture (0.7%, six of 848, p less than 0.001). There were no differences in clinical symptoms between patients with benign and malignant stricture. Seventeen of 165 benign strictures (10.3%) were long, extending over more than one anatomical segment of colon, but all 10 malignant strictures were short (p less than 0.0001). The age at the diagnosis of stricture was higher in the nine patients with malignant stricture than in the 123 patients with benign stricture (mean age 57.2 vs. 41.4 yr, respectively, p less than 0.01). The proportion of strictures that were malignant increased with duration of disease from 3.3% with less than 20 yr of CD, to 11% with CD of 20 yr or more. All nine patients with malignant stricture were treated surgically, and four of the nine died of colon cancer during a mean follow-up of 4.3 yr. Prognosis was worse in six other nonstricture cancers in this series, with five colon cancer deaths during mean follow-up of 1.6 yr. In view of the high rate of malignancy, 6.8% in this series, colonoscopy with biopsy is essential in Crohn's disease patients with colonic strictures, and surgery must be considered when a stricture cannot be fully assessed during colonoscopy.     

Fibrinopeptide A reactive peptides and procoagulant activity in bronchoalveolar lavage: relationship to rheumatoid interstitial lung disease

Extravascular, primarily, alveolar fibrin deposition is commonly associated with the alveolitis of many interstitial lung diseases including the interstitial lung disease associated with rheumatoid arthritis (RA). We therefore hypothesized that coagulation pathways, which promote fibrin formation, would be activated in the alveolar lining fluids of patients with rheumatoid interstitial lung disease. To test this hypothesis, we studied the bronchoalveolar lavage (BAL) fluids from patients with rheumatoid interstitial lung disease (n = 7) and patients with RA unassociated with interstitial lung disease (n = 10) to characterize and quantitatively compare the BAL procoagulant material and levels of fibrinopeptide A (FPA), which is cleaved from fibrinogen by thrombin. FPA reactive peptide concentrations were significantly greater in rheumatoid interstitial lung disease than RA when normalized per ml of concentrated BAL fluid (p = 0.02), per mg BAL total protein (p = 0.01) or BAL albumin content (p = 0.03) and correlated with BAL antigenic neutrophil elastase concentrations (r = 0.87). Procoagulant activity was present in similar concentration of BAL of patients with RA and rheumatoid interstitial lung disease and was mainly attributable to tissue factor associated with factor VII (or VIIa). Our results demonstrate that tissue factor and factor VII are endogenous in the alveoli of subjects with RA and interstitial lung disease and could interact with distal coagulation substrates which may enter the alveoli in interstitial lung disease to locally promote fibrin deposition.(ABSTRACT TRUNCATED AT 250 WORDS)     

Fibrinopeptide A and B beta 15-42 in liver cirrhosis

In order to detect even minimal fibrinolysis activation in liver cirrhosis, we measured fibrinopeptide B beta 15-42 (B beta 15-42), an indicator of plasmin activity in vivo and alpha 2-antiplasmin (alpha 2-AP) in a group of cirrhotic patients. The second goal of this study was to investigate whether an increased fibrinolytic activity is related to a chronic disseminated intravascular coagulation. For this purpose we concomitantly measured fibrinopeptide A (FPA), marker of thrombin activity in vivo. Results show significantly higher levels of B beta 15-42 in cirrhotic patients than in control (p less than 0.01). In patients with high FPA levels we found significantly higher values than in patients with normal FPA (p less than 0.01). alpha 2-AP was lower in patients with high FPA levels than in patients with normal FPA (p less than 0.05). A significant negative correlation was found between FPA and alpha 2-AP only in patients with high FPA (p less than 0.05). There was no relationship between B beta 15-42 and FPA nor between B beta 15-42 and alpha 2-AP when all patients were considered. These findings confirm that in liver cirrhosis fibrinolysis activation may occur. The primary pathogenetic role of DIC may be important in this respect. However the lack of correlation between FPA and B beta 15-42 suggests that other pathogenetic factors may be involved in determining fibrinolysis activation.     

Relationship between fibrinopeptide a (FPA) level and fibrinogen kinetics in patients with malignant disease

The relationship between FPA level and fibrinogen turnover rate as well as fibrinolytic activity was studied in 18 patients with malignant diseases. It was found that the FPA levels were significantly elevated and were correlated with fibrinogen turnover rate (r = 0.74, p less than 0.001) and FDP (r = 0.58, p less than 0.02). The estimated FPA turnover rate was also correlated with fibrinogen turnover rate (r = 0.70, p less than 0.001). These results suggest that fibrinogen catabolism in patients with malignant disease is related to thrombin proteolysis. However, ratios of 1/2 FPA turnover rate to fibrinogen turnover rate suggest that intravascular thrombin proteolysis is not the major determinant of fibrinogen catabolism. It is suspected that extravascular thrombin proteolysis is responsible for the elevation of plasma FPA level which is correlated with acceleration of fibrinogen catabolism.     

Association between blood rheology, thrombosis and cancer survival in patients with gynecologic malignancy

In cancer patients impaired blood rheology in the presence of coagulation activation may reduce blood flow in the vascular microcirculation that favors thrombosis but may also support tumor progression and metastasis. In 451 patients with gynecological cancer and 177 patients with corresponding benign tumor disease preoperatively, during adjuvant treatment, when venous thrombosis (VT) or cancer progression was diagnosed hematocrit (micro centrifuge), hemoglobin, leukocytes, platelets (Coulter Counter); red blood cell (RBC) aggregation (aggr.) during stasis and low shear conditions (MA 1, Myrenne), plasma viscosity (viscosimeter KSPV 1 Fresenius), and fibrinogen (Multifibren Behring Dade) were investigated. One hundred and twelve healthy women served as controls. Preoperatively, mean plasma viscosity (pv) was significantly higher in cancer patients as compared to patients with the corresponding benign tumor disease (breast cancer: n = 261; pv = 1.32 vs. 1.27 mPa s; p = 0.023; ovarian cancer: n = 68; pv = 1.39 vs. 1.31 mPa s; p < 0.001; endometrial cancer: n = 70; pv = 1.37 vs. 1.25 mPa s; p < 0.001; cervical cancer: n = 52; pv = 1.33 vs. 1.26 mPa s; p = 0.004). RBC aggr. was significantly lower in controls compared to the preoperative values in cancer patients but mean (median) values (RBC aggr. stasis < 21) were within the normal range in all. Preoperatively, plasma viscosity was a significant risk factor for the overall survival in ovarian cancer patients (p = 0.02) and for subsequent thrombosis in ovarian (p = 0.02) and cervical cancer patients (p = 0.007). In the multivariate analysis plasma viscosity was an independent prognostic marker for the overall survival of breast cancer patients (r = 99.45; 95% CI: 7.32-980.2; p < 0.0001). An optimized preoperative cut-off value above 1.40 mPa s (Log-Rank-test) was significantly associated with poor outcome in the Kaplan-Mayer survival estimates, even in node-negative breast cancer. In gynecologic cancer patients the combination of an increase in RBC aggregation and plasma viscosity impairs blood-flow-properties and may induce hypoxia in the microcirculation that favors thrombosis, settlement of tumor-cells and thus metastasis. Improvement of blood fluidity and thus oxygen transfer in the tumor-vascular-microcirculation may increase susceptibility of systemic anti-cancer therapy.     

Association between time of increased fibrinopeptide A levels in plasma and episodes of spontaneous angina: a controlled prospective study

Thirty-seven patients affected by spontaneous angina and 15 comparable control subjects were enrolled in a 12-month prospective study to evaluate the relationship between blood clotting activation (assessed by fibrinopeptide A [FPA] plasma concentration) and the occurrence of myocardial ischemic attacks. FPA measurements and clinical examinations in patients were performed every 2 weeks. In control subjects blood sampling was performed every 4 weeks. Data from 28 patients who completed the study and from the 15 control subjects were analyzed. The clinical activity of angina was divided into three classes (asymptomatic, mildly symptomatic, and severely symptomatic) on the basis of the number and time-concentration of the ischemic attacks and ECG changes during the 15 days preceding each clinical examination. In all but one patient, a cyclic pattern of activity of coronary artery disease was observed. During follow-up studies, 624 FPA measurements were performed in patients and 173 in control subjects. Mean values were 4.68 +/- 4.53 and 1.32 +/- 0.60 ng/ml, respectively (p less than 0.001). FPA levels differed markedly in relation to the activity of angina. A relationship between FPA levels and activity of disease (r = 0.54, p less than 0.01) was found in time course. Bolus heparin administration (100 IU/kg) during the active phase of angina sharply but incompletely lowered FPA plasma levels, indicating thrombin formation both intravascularly and extravascularly. Present results indicate that a marked blood clotting activation occurs simultaneously with the outbursts of clinical activity of spontaneous angina.     

Is sickle cell crisis a thrombotic event?

Although large vessel thrombi are occasionally reported in patients with homozygous sickle cell disease, the role of intravascular coagulation in typical pain crises is controversial. Therefore, we studied 24 sickle cell patients during and between episodes of pain crisis, using several sensitive tests of hemostasis. Fibrinogen was measured by a clotting assay, beta-thromboglobulin (beta-TG) and fibrinopeptide A (FPA) were quantitated by radioimmunoassay, and protein C was determined by absorbing the zymogen from test plasma, activating it with thrombin-thrombomodulin complex, and measuring activity with a selective synthetic substrate. Fibrinogen was elevated in asymptomatic patients (355 +/- 145 mg/dl) but was no different from the value in these same patients during crisis (333 +/- 180 mg/dl, p greater than 0.1). Similarly, beta-TG 136 +/- 52 ng/ml vs 118 +/- 56; FPA 3.7 +/- 4.8 ng/ml vs 5.2 +/- 4.5, and protein C 71 +/- 20% vs 66 +/- 19 showed no important changes during crisis. However, all these values were significantly different from those in age- and sex-matched healthy controls. beta-TG, fibrinogen, and FPA were elevated (p less than 0.001, 0.005, and 0.05, respectively), and protein C was decreased (p less than 0.003). We conclude that while chronic intravascular coagulation is common in patients with sickle cell disease, there is no evidence that the pain crisis per se is a thrombotic event.     

Cholecystectomy and colon cancer

A hospital-based case-control study on the possible relationship between cholecystectomy and colon cancer is described. In all, 569 cases and 1,129 controls were included in the study. Controls were patients who had been admitted either with another malignant disease (lung cancer or breast cancer), or because of benign gastrointestinal diseases. No differences with regard to colon cancer risk were observed between the two control groups. The crude odds ratio (OR) after cholecystectomy for colon cancer was 2.1 [95% CL (confidence limits) = 1.5; 2.9]. No significant difference in risk between males (OR = 2.0; 95% CL = 1.1; 3.6) and females (OR = 2.1; 95% CL = 1.4; 3.2) was found. A significant inverse relationship was found between the relative risk for colon cancer and sublocalization of the colon cancer, the highest relative risk (RR) found proximal in the colon. A decreasing risk with increasing interval between cholecystectomy and date of diagnosis of colon cancer was found. It is hypothesized that the apparent association between cholecystectomy and colon cancer does not reflect a necessary cause-effect relationship, but is rather an epiphenomenon representing a diminished exposure level to other etiologic factors.     

Relationship between elevated fibrinopeptide A levels and alpha-2-antiplasmin

In order to investigate whether alpha 2-antiplasmin (alpha 2-AP) levels may be related to thrombin activity, we measured alpha 2-AP and fibrinopeptide A (FPA) in 51 patients with clinical conditions frequently associated with increased thrombin activity. The diagnoses were: atherosclerotic disease, chronic inflammatory disease and hematological neoplastic disease. A significant negative correlation was found between alpha 2-AP and FPA (p less than 0.01). When patients were divided into three subgroups on the basis of their FPA levels, a significant reduction in alpha 2-AP was found in patients with the highest FPA concentration (greater than 9 ng/ml). Accordingly, a significant negative relationship between alpha 2-AP and FPA was found only in this subgroup (p less than 0.01). Our data suggest that the partial consumption of alpha 2-AP in patients with elevated FPA levels may reflect a subclinical fibrinolysis activation secondary to increased thrombin activity.     

Disturbed fibrinolysis in patients with inflammatory bowel disease. A study in blood plasma, colon mucosa, and faeces.

Using specific assays, we studied fibrinolytic activity in plasma and colonic mucosa biopsies of 28 patients with inflammatory bowel disease (IBD) (12 with Crohn's disease, 16 with ulcerative colitis) and 28 control patients without inflammatory bowel disease or colon malignancy. Blood coagulation was studied using standard techniques. In plasma of IBD patients significantly decreased tissue type plasminogen activator activity (t-PA) (p less than 0.02), increased plasminogen activator inhibition (PAI) (p less than 0.01) and fibrinogen (p less than 0.001), and prolonged thrombin time (p less than 0.001) and prothrombintime (p less than 0.001) were found. In colon mucosa the percentage of t-PA to urokinase type plasminogen activator (u-PA) was 80:20% in the control group and 71:29% in the IBD group in non-inflamed mucosa. In inflamed mucosa the plasminogen activator percentage was 55:45%, significantly different (p less than 0.01) from the control group. There was also a significant absolute increase in u-PA activity and decrease of t-PA activity in the inflamed mucosa compared to the control group (p less than 0.001 and p less than 0.01, respectively). Patients with inflammatory bowel disease therefore have significant changes in components of the fibrinolytic and coagulation system both systemically and locally in colon mucosa. These changes might contribute to an increased risk for thromboembolic complications and possibly to the pathogenesis of the colitis and to the local complications such as bleeding.     

Plasma concentration of platelet-specific proteins and fibrinopeptide A in patients with artificial heart valves

beta-Thromboglobulin (beta TG), platelet factor 4 (PF4), fibrinopeptide A (FPA), lactic dehydrogenase (LDH), and platelet count were evaluated in patients with bioprostheses and prosthetic heart valves. beta TG and PF4 were significantly elevated in both patient groups (p less than 0.001), whereas FPA was normal. There was no significantly difference in plasma concentrations of beta TG and PF4 between patients with prosthetic heart valves and bioprostheses. LDH levels were significantly (p less than 0.001) higher and platelet count lower (p less than 0.001) in patients with prosthetic cardiac valves. The data indicate that bioprostheses do not cause haemolysis or activation of the coagulation system. The findings that the plasma concentration of platelet-specific proteins were elevated, support the assumption that both types of valves cause platelet damage.     

Recurrence patterns after curative resection of colorectal cancer in patients followed for a minimum of ten years

BACKGROUND/AIMS: To investigate the recurrence patterns and interval from initial surgery in patients with curatively resected colorectal cancer followed for a minimum of 10 years. METHODOLOGY: We retrospectively reviewed 418 patients who had undergone curative resection for colon cancer (n = 246) or rectal cancer (n = 169). Follow-up periods ranged from 10 to 23 years. Main outcome measures were interval until recurrence, site of first recurrence, and influence of adjuvant chemotherapy. RESULTS: 26 (6%) had been lost to follow-up by 10 years and 143 (34%) had died. The most common site of recurrence was liver in colon cancer and locoregional in rectal cancer. The cumulative recurrence rate in colon cancer was 100% at 4 years. In rectal cancer, it was 89% at 5 years, 98% at 7 years and 100% at 10 years. The interval until recurrence was longer in rectal cancer (26.0 +/- 24.2 months) than in colon cancer (17.1 +/- 11.0 months) (p = 0.03). It was also longer in patients receiving than in those not receiving adjuvant chemotherapy (p < 0.01). The interval until lung metastasis was longer than that until liver metastasis in colon cancer (p = 0.04), and longer than that until locoregional recurrence in rectal cancer (p = 0.03). The interval until recurrence in the colon cancer was shorter for stage III than for stage II (p = 0.02). CONCLUSIONS: Surveillance for recurrences, particularly for relapses in the liver and lung, should be performed for at least 4 years in colon cancer patients. Patients with rectal cancer should be followed for a longer period than those with colon cancer, focusing on locoregional, liver and lung recurrence. It is particularly noteworthy that adjuvant chemotherapy may prolong the interval until recurrence and the interval until lung metastasis is relatively longer.     

Interleukin-1 and interleukin-6 in breast cyst fluid: their role in regulating aromatase activity in breast cancer cells.

Gross cystic breast disease is a common benign disease which may be associated with an increased risk for breast cancer. Breast cyst fluid (BCF) contains many steroids, peptide growth factors and proteins. We have now identified interleukin-1 (IL-1) and IL-6 in BCF by specific radioimmunoassays. Concentrations of IL-1 were similar in BCF with low or high Na+/K+ ratios (ratio less than 3 vs greater than 3; 357 +/- 72 pg/ml vs 308 +/- 126 pg/ml). In contrast, IL-6 concentrations were significantly higher (P less than 0.01) in BCF with a Na+/K+ ratio greater than 3 (2.75 +/- 2.34 ng/ml) compared with BCF with a low electrolyte ratio (0.21 +/- 0.09 ng/ml). BCF (10%, v/v) stimulated aromatase activity when added to dexamethasone stimulated breast tumour-derived fibroblasts and there was a significant correlation between the stimulation of aromatase activity and BCF Na+/K+ ratio (r = 0.95, P less than 0.001). A significant correlation was also found between stimulation of aromatase activity and concentration of IL-6 in BCF (r = 0.80, P less than 0.01) but not IL-1 concentration (r = -0.39, not significant). Addition of IL-1 or IL-6 (50 ng/ml) to fibroblasts stimulated aromatase activity but was associated with a small (20%) decrease in cell growth. It is concluded that IL-6 may have an important role in regulating aromatase activity in breast cancer cells.     

The relationship between the clinical presentation and spread of colon cancer in 315 consecutive patients. A significant trend of earlier cancer detection from 1982 through 1988 at a university hospital.

We systematically analyzed the relationship between 47 clinicoepidemiologic parameters and stage of colon cancer in 315 patients who underwent colon cancer surgery from 1982 through 1988 at the Robert Johnson University Hospital. A history of hemorrhoids was correlated with early cancer, possibly because of earlier self-referral (odds ratio = 18.2; chi 2 = 10.4; degrees of freedom = 1; p less than 0.001). However, anemia was correlated with advanced cancer (odds ratio = 0.21; chi 2 = 13.7; degrees of freedom = 1; p less than 0.0002). Anemia may result from chronic bleeding due to a longstanding cancer. Prior studies have suggested that intensive screening programs may produce earlier colon cancer detection; this study demonstrated for all patients at a medical center a significant trend from 1982 through 1988 of detecting colon cancer at an earlier pathologic stage and with a better differentiated histologic grade (for first half of study period 44.4% had Dukes' stage A or B cancer, second half of study period 58.6% had Dukes' stage A or B cancer; odds ratio = 0.56; chi 2 = 5.8; degree of freedom = 1; p less than 0.02). Possible explanations for this phenomenon are earlier self-referral because of increased patient awareness of cancer warning signs, and earlier physician detection because of greater use of colonoscopy and polypectomy and because of increased screening and surveillance. This earlier detection may herald a future significant decrease in colon cancer mortality at this hospital because prognosis is closely related to cancer stage. Further studies are required to determine if this is part of a national trend.     

Factors affecting fibrinopeptide-A levels in patients with venous thromboembolism during anticoagulant therapy

The prompt reduction of elevated fibrinopeptide A (FPA) levels (normal less than 1.3 pmole/ml) by heparin therapy in patients with thromboembolism suggests that measuring the FPA level may provide a good index of disease activity and be a useful method of monitoring therapy. Sepsis or malignancy may elevate FPA levels and coexist with thromboembolism. FPA levels were surveyed in 51 patients with thromboembolism (including 15 with concurrent sepsis or malignancy) during heparin treatment in an attempt to distinguish the effects of coexistent disease and the progression of thromboembolism. The anticoagulant effect of heparin was within the therapeutic range for 81% of the study period. In patients with thromboembolism alone and marked resolution of emboli on repeat lung scan, the mean daily FPA levels were lower than the values in patients with minimal resolution (p less than 0.005). In patients with marked resolution of pulmonary embolism or venous thrombosis and a concurrent disorder, the mean FPA level remained elevated compared to normal values in patients with thromboembolism alone. These results suggest that FPA levels monitored during heparin therapy of thromboembolism may be useful as an index of disease activity except in the presence of coexisting sepsis or malignancy.     

The effect of undernutrition and disease site and stage on cell-mediated immunity in gastrointestinal cancer

In order to assess whether impairment of cell-mediated immunity in cancer was due to the presence, site and stage of malignant disease or to associated undernourishment, 48 patients with gastrointestinal cancer were compared to 27 patients with benign gastrointestinal disease. Cell-mediated immunity (CMI) was assessed by delayed hypersensitivity (DHS) skin test response to recall antigens and lymphocyte mitogenic response to phytohaemagglutinin (PHA). Patients were grouped according to nutritional status assessed from a history of weight loss and changes in the ratio of total body potassium (TBK mmol) to total body water (TBW kg). The results showed that, whilst cancer patients as a whole had significantly worse DHS response, there were no significant differences between benign and malignant groups of equivalent nutritional status and in both types of disease response decreased significantly with worsening nutritional status (chi 2(2) = 6.83 cancer, 6.11 benign, p less than 0.05). PHA response was significantly worse in the worst nourished cancer patients than in those with normal nutritional status (p less than 0.05, Wilcoxon). Disease site and stage had no significant effect on either test. The results suggest that depression of CMI response in gastrointestinal cancer is due to undernutrition and is not a cancer specific effect.