The influence of whole-body vs. torso pre-cooling on physiological strain and performance of high-intensity exercise in the heat.

Little research has been reported examining the effects of pre-cooling on high-intensity exercise performance, particularly when combined with strategies to keep the working muscle warm. This study used nine active males to determine the effects of pre-cooling the torso and thighs (LC), pre-cooling the torso (ice-vest in 3 degrees C air) while keeping the thighs warm (LW), or no cooling (CON: 31 degrees C air), on physiological strain and high-intensity (45-s) exercise performance (33 degrees C, 60% rh). Furthermore, we sought to determine whether performance after pre-cooling was influenced by a short exercise warm-up. The 45-s test was performed at different (P<0.05) mean core temperature [(rectal+oesophageal)/2] [CON: 37.3+/-0.3 (S.D.), LW: 37.1+/-0.3, LC: 36.8+/-0.4 degrees C] and mean skin temperature (CON: 34.6+/-0.6, LW: 29.0+/-1.0, LC: 27.2+/-1.2 degrees C) between all conditions. Forearm blood flow prior to exercise was also lower in LC (3.1+/-2.0 ml 100 ml tissue(-1) x min(-1)) than CON (8.2+/-2.5, P=0.01) but not LW (4.3+/-2.6, P=0.46). After an exercise warm-up, muscle temperature (Tm) was not significantly different between conditions (CON: 37.3+/-1.5, LW: 37.3+/-1.2, LC: 36.6+/-0.7 degrees C, P=0.16) but when warm-up was excluded, T(m) was lower in LC (34.5+/-1.9 degrees C, P=0.02) than in CON (37.3+/-1.0) and LW (37.1+/-0.9). Even when a warm-up was performed, torso+thigh pre-cooling decreased both peak (-3.4+/-3.8%, P=0.04) and mean power output (-4.1+/-3.8%, P=0.01) relative to the control, but this effect was markedly larger when warm-up was excluded (peak power -7.7+/-2.5%, P=0.01; mean power -7.6+/-1.2%, P=0.01). Torso-only pre-cooling did not reduce peak or mean power, either with or without warm-up. These data indicate that pre-cooling does not improve 45-s high-intensity exercise performance, and can impair performance if the working muscles are cooled. A short exercise warm-up largely removes any detrimental effects of a cold muscle on performance by increasing Tm.

Effect of pre-cooling, with and without thigh cooling, on strain and endurance exercise performance in the heat.

Body cooling before exercise (i.e. pre-cooling) reduces physiological strain in humans during endurance exercise in temperate and warm environments, usually improving performance. This study examined the effectiveness of pre-cooling humans by ice-vest and cold (3 degrees C) air, with (LC) and without (LW) leg cooling, in reducing heat strain and improving endurance performance in the heat (35 degrees C, 60% RH). Nine habitually-active males completed three trials, involving pre-cooling (LC and LW) or no pre-cooling (CON: 34 degrees C air) before 35-min cycle exercise: 20 min at approximately 65% VO2peak then a 15-min work-performance trial. At exercise onset, mean core (Tc, from oesophagus and rectum) and skin temperatures, forearm blood flow (FBF), heart rate (HR), and ratings of exertion, body temperature and thermal discomfort were lower in LW and LC than CON (P<0.05). They remained lower at 20 min [e.g. Tc: CON 38.4+/-0.2 (+/-S.E.), LW 37.9+/-0.1, and LC 37.8+/-0.1 degrees C; HR: 177+/-3, 163+/-3 and 167+/-3 b.p.m.), except that FBF was equivalent (P=0.10) between CON (15.5+/-1.6) and LW (13.6+/-1.0 ml.100 ml tissue(-1) x min(-1)). Subsequent power output was higher in LW (2.95+/-0.24) and LC (2.91+/-0.25) than in CON (2.52+/-0.28 W kg(-1), P=0.00, N=8), yet final Tc remained lower. Pre-cooling by ice-vest and cold air effectively reduced physiological and psychophysical strain and improved endurance performance in the heat, irrespective of whether thighs were warmed or cooled.

Aspirin effects on endometrial cancer cell growth.

OBJECTIVE: To find whether aspirin (acetylsalicylic acid, ASA) inhibits the growth of endometrial cancer cells in vitro in a way similar to that in colorectal cancer cells and to investigate the mechanisms by which aspirin might lead to growth inhibition. METHODS: Ishikawa human endometrial tumor cells were grown in the presence of ASA (1-5 mM) for 96 hours. Controls were treated with vehicle (absolute ethanol). Cell proliferation was assessed by 3-(4,5-dimethylthiazol-2-yl) -2,5-diphenyltetrazolium bromide assay. Apoptosis was determined by terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling assay. Analysis of cell-cycle distribution and bcl-2 expression was assessed by flow cytometry. RESULTS: Acetylsalicylic acid induced a dose-dependent inhibition of Ishikawa cells in vitro. The percentage of growth inhibition was 21-88% at concentrations of 1-5 mM. It also induced apoptosis and reduced bcl-2 expression in Ishikawa cells in a dose-dependent manner. Control cells and cells treated with the lowest concentration of ASA exhibited 2% apoptosis and more than 60% of the population expressed bcl-2. Apoptosis levels increased as levels of ASA increased from 2 to 5 mM (7-58%) with a concommitant decrease in bcl-2 expression from 46% at 2 mM to 2% at 5 mM. Acetylsalicylic acid concentrations of 3 mM or greater induced a shift from the resting phase (G0/G1) to S phase of the cell cycle. CONCLUSION: Acetylsalicylic acid inhibited Ishikawa cell growth in vitro in a dose-dependent manner. Apoptosis is one of the mechanisms involved in the response, which can be mediated in part by downregulation of bcl-2.

Accuracy of lymph node palpation to determine need for lymphadenectomy in gynecologic malignancies.

OBJECTIVE: To assess the accuracy of intraoperative lymph node palpation for identifying lymph node metastasis in gynecologic malignancies. METHODS: We prospectively evaluated 126 women who had lymphadenectomies for staging of various gynecologic malignancies from August 1995 to June 1997. Surgeries were done by obstetrician-gynecologists with subspecialty certification in gynecologic oncology from the American Board of Obstetrics and Gynecology, who had practiced gynecologic oncology for at least 5 years. Data were collected on gynecologic oncologists' opinions of lymph node status by palpation. Nodes believed to be positive were sent separately. We recorded operating time for lymphadenectomies, and intraoperative and postoperative complications. RESULTS: Mean (range) patient age was 55 (18-83) years. Mean (range) operating time was 188 (85-435) minutes. The mean (range) lymphadenectomy time was 46 (5-150) minutes. The total number of lymph nodes dissected was 2138. One hundred seven of 2138 (5%) nodes were positive for malignancy. Thirty-eight of 107 (36%) positive lymph nodes were missed by palpation. Fifty-six of 2031 (3%) negative lymph nodes were selected as positive. Sixty-nine of 107 (64%) positive lymph nodes were identified correctly. Sensitivity and specificity of palpation were 72% and 81%, respectively. The positive and negative predictive values of lymph node palpation were 56% and 89%, respectively. CONCLUSION: Intraoperative lymph node palpation has low sensitivity and positive predictive value even when done by experienced board-certified gynecologic oncologists.

Detection of sentinel lymph nodes with lymphazurin in cervical, uterine, and vulvar malignancies.

BACKGROUND: Identification of sentinel lymph nodes may allow prediction of metastatic disease in cancer patients. We did a prospective study to determine whether lymphazurin dye could identify sentinel lymph nodes in patients with cervical, uterine, and vulvar cancer. METHODS: In 33 patients having surgery for either uterine, cervical, or vulvar carcinoma, lymphazurin dye was injected into the respective organs before the tumor and node dissection began. Sentinel lymph nodes were identified and dissected in situ. RESULTS: The identification rate of sentinel lymph nodes was 0/8 (0%) for uterine cancer patients, 2/13 (15.4%) for cervical cancer patients, and 9/12 for vulvar cancer patients (75%). CONCLUSIONS: In a limited number of patients, lymphazurin day may be useful in identifying or assessing the sentinel nodes draining vulvar and cervical cancers. The role of this procedure in treatment planning for patients with gynecologic malignancies is yet to be determined.

Regional blood flow response to hypothermia in premature, newborn, and neonatal piglets.

BACKGROUND/PURPOSE: Hypothermia (HT) remains a significant stress to the newborn and has been implicated in the pathogenesis of necrotizing enterocolitis (NEC). The authors assessed the effect of transient HT (32 degrees C) on regional organ blood flow in anesthetized piglets at age 7 to 10 days preterm (PREM), 1 to 2 days (NB), and 1 to 2 weeks (NEO). METHODS: Radiolabeled microspheres were used to determine organ blood flows (mL/min/g) at baseline, 15, and 60 minutes after HT and 60 minutes after rewarming to baseline core temperature. RESULTS: Heart rate and cardiac output decreased significantly in all groups. Cardiac flow decreased significantly in the NEO group, and central nervous system (CNS) flow decreased significantly in the NB and NEO groups. Both returned to baseline levels after rewarming. The PREM group experienced decreased cardiac, CNS, and intestinal blood flows but not to significant levels. NB and NEO intestinal blood flow showed significant decreases, which remained so after rewarming (a response not seen in hypoxia or hypovolemia). Cardiac output did not return to baseline levels in any group. CONCLUSIONS: HT causes derangements in organ blood flows that differ from other deleterious stimuli such as hypoxia and hypovolemia. The prolonged intestinal ischemia supports HT as a factor in the development of NEC. This delay may offer opportunity to intervene in an attempt to lessen ischemia-reperfusion injury.

Does ligation of the hypogastric artery at the time of radical hysterectomy and lymphadenectomy decrease blood loss? Results of a prospective randomized trial.

A prospective, randomized study of patients undergoing radical hysterectomy for gynecologic malignancies was undertaken from 10/95 to 11/96 to determine if ligation of the hypogastric arteries at the time of radical hysterectomy decreases blood loss. Patients were randomized to either ligation of the hypogastric artery (Group 1) or no ligation (Group 2) prior to a standard Piver type III radical hysterectomy. Surgeries were performed by Board certified gynecologic oncologists with gynecologic oncology fellows and/or OB/GYN residents. Patients were analyzed for demographic characteristics and intraoperative and postoperative parameters. Statistical analysis was performed with independent samples t-test, Mann-Whitney rank sum test, Chi square and Fisher exact test. Twenty-one patients were randomized to group 1 and 22 to group 2. Groups were similar with respect to demographics and preoperative parameters except for age. There were no differences among the groups with respect to intraoperative and postoperative parameters. The mean estimated blood loss for group 1 was 600 ml and 550 ml for group 2 (P = NS). Hypogastric artery ligation (HAL) at the time of radical hysterectomy for gynecologic malignancy does not reduce blood loss.

Body mass predicts the survival of patients with new International Federation of Gynecology and Obstetrics Stage IB1 and IB2 cervical carcinoma treated with radical hysterectomy.

BACKGROUND: The authors evaluated the impact of body mass on survival and morbidity of patients with new International Federation of Gynecology and Obstetrics (FIGO) Stage IB1 and IB2 cervical carcinoma managed with radical hysterectomy. METHODS: Two hundred twenty-nine patients with Stage IB1 or IB2 cervical carcinoma treated with radical hysterectomy were studied in a multivariate logistic regression analysis. The body mass index (BMI) and the ponderal index (PI) were used as measures of body mass and were analyzed as predictors of recurrence, survival, and complications in light of the new staging system. RESULTS: Twenty-seven of 229 patients died of recurrent disease. A low BMI or a high PI were predictive of poor survival. Tumor greatest dimension, lymph node involvement, BMI, and PI were all independent predictors of survival (P=0.0006). The only independent predictor of complications was para-aortic lymph node dissection (P=0.0026). CONCLUSIONS: Cervical carcinoma patients with a low body mass, as indicated by a low BMI or a high PI, were found to have poor survival after undergoing radical hysterectomy. Additional predictors of poor survival included lymph node metastases and increased tumor size. BMI and PI are more important predictors of survival than the new FIGO Stages IB1 and IB2. Body mass is not predictive of complications.

Organ blood flow redistribution in response to hypoxemia in neonatal piglets.

This study was designed to determine the effects of severe hypoxemia on newborn piglet visceral blood flow. While the hemodynamic effects of a severe hypoxemic insult are well characterized in newborn animals, its impact on organ perfusion in premature infants is not well characterized. Cannulas were placed in the femoral vessels and left atrium of term (1-14 days old) and prematurely delivered (cesarean section at 90% of term gestation) piglets. After stabilization, some animals were subjected to 1 h of ventilator-controlled hypoxia (yielding PaO2 approximately = 30-40 torr) followed by 30 min of reoxygenation; the remaining animals served as unchallenged controls. Radiolabeled microspheres were injected in all animals at times 0 min (baseline), 5 and 60 min (hypoxia), and 90 min (reoxygenation). Blood flows (mL/min/g tissue) to organs were determined using reference organ techniques. Control animals displayed no alterations in any of the variables monitored. Throughout the experimental period, organ blood flows were almost uniformly lower (p<.05, ANOVA) in premature versus term animals. The trend toward increased cerebral and cardiac blood flows during hypoxia observed in the premature piglets was similar to that of term animals, but of lower magnitude. In term piglets, hypoxia produced an immediate and significant (*p<.05) decline in small-intestinal blood flow followed by autoregulatory escape (2.02+/-0.17 mL/min/g at time 0, 1.56+/-0.15 mL/min/g at 5 min hypoxia, 1.88+/-0.18 mL/min/g at 60 min hypoxia, 2.26+/-0.19 mL/min/g at 30 min reoxygenation), an effect not readily observed in the premature piglets (0.48+/-0.10 mL/min/g at time 0, 0.44+/-0.07 mL/min/g at 5 min hypoxia, 0.46+/-0.10 mL/min/g at 60 min hypoxia, 0.42+/-0.08 mL/min/g at 30 min reoxygenation). However, mucosal blood flows measured in these younger animals declined throughout the experimental period to almost 50% of baseline, compared to a complete restoration to baseline blood flow observed following reoxygenation of term piglets. Intestinal blood flow in premature infants is small when compared to term animals, and alterations in small intestinal blood mucosal flow induced by hypoxia appear less well tolerated by the premature animals. Taken together, this may in part account for the increased risk of developing intestinal ischemic diseases in premature infants who are even temporarily exposed to a severe hypoxic challenge.

A pilot study utilizing intraoperative lymphoscintigraphy for identification of the sentinel lymph nodes in vulvar cancer.

OBJECTIVE: To identify sentinel lymph nodes using intraoperative lymphoscintigraphy. METHODS: Technetium-99-labeled sulfur colloid was injected at the site of primary vulvar carcinoma. An intraoperative gamma counter was used to identify one or more sentinel lymph nodes. RESULTS: Ten patients underwent bilateral inguinal and femoral lymphadenectomy. The clinical stages are as follows: T1 in 6, T2 in 2, and T3 in 2. A total of four groins (3 patients) were positive for metastases. In one patient only the sentinel node was positive for disease. In a second patient, two unilateral nodes were positive for disease and both were identified with the gamma counter as sentinel nodes. In the third patient, a single sentinel node was positive for malignancy in each groin. Multiple nonsentinel lymph nodes were positive in each groin in this patient. In no case was the sentinel node negative when other nonsentinel nodes were positive. CONCLUSION: Intraoperative lymphoscintigraphy quantitatively identifies one or more sentinel lymph nodes. Since sentinel lymph nodes can be localized transcutaneously, this technique may be useful for selective lymphadenectomy. Larger patient accrual is necessary to verify this technique.

Management of gynecologic oncology patients with a preoperative deep vein thrombosis.

Our experience with gynecologic oncology patients presenting preoperatively with a deep vein femoral thrombosis is reported. Over a 3-year period data were collected on all patients at the University of South Florida (USF) requiring surgery for a known or suspected gynecologic cancer and having a concomitant active femoral venous thrombosis. Twelve such patients were managed. Management was divided among three options: heparinization, preoperative inferior vena cava (IVC) filter, and intraoperative IVC ligation. For two patients a filter could not be placed preoperatively due to tumor compression of the IVC. Both underwent IVC ligation intraoperatively. One of the two died intraoperatively, possibly related to pulmonary embolism. One of eight with a preoperative IVC filter had obvious clot propagation postoperatively, managed with heparin. One of two managed with heparin only had severe bleeding and heparin-associated thrombocytopenia (HAT) preoperatively. Based on our experience and a review of the literature, we recommend therapeutic heparinization and a preoperatively placed IVC filter for most preoperative gynecologic oncology patients with femoral deep venous thrombosis.

The clinical significance of calculated versus measured creatinine clearance in patients with gynecologic malignancy.

Measurement of creatinine clearance remains in convenient because of the need for 24-hour urine collection. Our study compared creatinine clearance calculated from serum creatinine (CC alpha) to measured creatinine clearance (MC alpha) using an unsupervised 8-hour urine collection at home. Forty-two women admitted to our gynecologic oncology service participated in this study. Twenty-six of these patients had cancer originating from the ovaries, 8 from the uterus, 7 from the cervix, and 1 from the fallopian tube. There were adequate data for 61 collections. MC alpha using unsupervised 8-hour urine collection at home correlated well with CC alpha. This calculated value may be sufficient to be used as a guide in evaluating renal function in patients with gynecologic malignancy. However, if it becomes necessary to measure creatinine clearance, our data suggest that an 8-hour collection may be used.

Massive pelvic hemorrhage during gynecologic cancer surgery: "pack and go back".

Over a 13-year period, intraabdominal packing has been used to control massive hemorrhage during surgery for gynecologic malignancy in six patients. Five patients had undergone total pelvic exenteration and one total abdominal hysterectomy with bilateral salpingo-oophorectomy for endometrial cancer. Massive hemorrhage was defined as infusion of more than 10 units of blood and replacement of more than one total blood volume. Tamponade was performed using continuous Kerlex rolls (Kendall Co., Boston, MA) in a bowel bag with directed pressure over the hemorrhaging site with abdominal closure. The packs were removed in 48 to 72 hr in the operating room, transabdominally in five patients and transvaginally in one. One postoperative death occurred within 8 hr of surgery. The packing was ultimately successful in the five remaining patients. In five of six patients, tumors were removed before the packing, whereas in one, the tumor was removed concurrently with the pack. In one patient, immediate repacking was required after pack removal, with ultimate hemostasis. Morbidity included "empty pelvis syndrome" in four patients, neuropathy in three (obturator in 1, sciatic in 2), and small bowel obstruction in one. In patients with severe intraoperative hemorrhage, intraabdominal packing has been successful as a mode of treatment.

Radical hysterectomy for stage IB1 vs IB2 carcinoma of the cervix: does the new staging system predict morbidity and survival?

Two hundred twenty-nine patients with Stage IB cervical cancer treated with radical hysterectomy were assigned to the new FIGO substages IB1 (n = 181) and IB2 (n = 48) based on clinical tumor diameter. Our purpose was to determine the impact of the new staging system for IB1 and IB2 cervical cancer on nodal status and survival. Additionally, we analyzed the morbidity of radical hysterectomy in light of the new staging system. The complications were similar between the two groups. Para-aortic lymphadenectomy was the only independent predictor of complications (P = 0.00026). Stage IB2 patients did have a significantly worse 5-year survival (72.8%) when compared with IB1 (90.0%) (P = 0.0265). Multivariate stepwise logistical regression analysis indicated that the new staging system did not have an independent impact on survival. Stage acts through nodal status in its impact on survival. Positive lymph nodes, tumor diameter, and Ponderal Index are all independent predictors of survival (P = 0.0001). Patients with Stage IB2 carcinoma of the cervix undergoing radical hysterectomy showed no significant increase in morbidity when compared with patients with Stage IB1 disease treated with the same procedure.

Intraoperative liver biopsy with the loop electrosurgical excision procedure in patients with gynecologic malignancies.

When patients with gynecologic malignancies undergo exploratory surgery, hepatic lesions suggestive of malignancy are occasionally encountered. In 18 patients undergoing laparotomy for gynecologic cancer, visible liver lesions suggestive of malignancy were biopsied with the loop electrosurgical excision procedure (LEEP). Hemostatic suture placement was required in 3 of these 18 patients, and in 15, hemostasis was achieved with electrocautery only. Three of the 18 biopsies (16.7%) were positive for malignancy. No intraoperative or postoperative complications were attributed to the liver biopsy. One patient required postoperative transfusion of 2 units packed red blood cells. A new technique is described using the LEEP to remove suspicious lesions during exploratory laparotomy in patients with gynecologic malignancies.

Radical hysterectomy for stage IB cervical cancer: recurrence interval as a predictor of survival.

Two hundred patients with stage IB carcinoma of the cervix were treated with radical hysterectomy. We evaluated the impact of recurrence interval on survival and determined its relationship to other prognostic factors. Thirty-one patients (15.5%) had recurrence, with a median follow-up of 2.8 years (range, 1 to 5 years). Multivariate analysis, using the Cox proportional hazard regression model, showed the impact of recurrence interval on survival and its relationship to other prognostic factors. Patients were 19 times more likely to die during follow-up if recurrence occurred shortly after the operation. However, the risk of death from recurrence decreased exponentially as recurrence interval increased, by a multiple of 0.93 m where m is recurrence interval in months. We conclude that in patients with stage IB carcinoma of the cervix treated initially with radical hysterectomy, the shorter the recurrence period after operation, the greater the likelihood the patient would die during 5-year follow-up. This information may help clinicians determine a patient's prognosis after confirmed recurrence.

Chemotherapy plus sequential hormonal therapy for advanced and recurrent endometrial carcinoma: a phase II study.

We evaluated the therapeutic value of sequential cyclical hormonal therapy (megestrol acetate, and tamoxifen citrate) plus single-agent chemotherapy (carboplatin) in the outpatient management of advanced or recurrent endometrial cancer. Carboplatin (300 mg/m2) was administered every 4 weeks for six courses or until disease progression. In addition, patients alternated megestrol acetate (80 mg orally twice daily) with tamoxifen citrate (20 mg orally twice daily) every 3 weeks. Thirteen of 18 (72.2%) patients were considered evaluable. Four patients (30.8%) had a complete response, six (46.2%) had a partial response, one (7.7%) had stable disease, and two patients (15.4%) progressed. Six of seven patients with vaginal disease responded. The median progression-free interval was 14 months for complete responders. Two patients (15.4%) are alive with no evidence of disease at 41 and 59 months. Seven of 13 patients experienced a hematologic toxicity (six grade 2, one grade 3); all resolved within 2 weeks. Dose reduction of carboplatin to 200 mg/m2 was required in one patient. No other toxicities were encountered. The median survival for all patients is 11 months, and is 33 months for complete responders. We conclude that a regimen of carboplatin plus sequential hormonal therapy shows promise in this pilot study for the treatment of advanced or recurrent endometrial cancer.

A single institution experience with weekly intramuscular methotrexate for nonmetastatic gestational trophoblastic disease.

The purpose of this study is to report our experience with weekly intramuscular methotrexate for nonmetastatic gestational trophoblastic disease. Treatment was begun at a dose of 40 mg/m2 and escalated weekly by 5-10 mg/m2, depending upon response and tolerance, to a maximum dose of 60 mg/m2. Remission was induced with weekly intramuscular methotrexate alone in 12 (60%) of 20 patients in 2-12 (median 8) weeks. The remaining 7 patients had a complete response to alternate chemotherapy. There were no major toxicities. Although the results of the present study are less favorable in terms of response, the overall published results using weekly intramuscular methotrexate for nonmetastatic gestational trophoblastic disease indicate that it is comparable in efficacy to other first-line treatments while having the advantages of convenience, low cost, and low toxicity.

Urokinase plasminogen activator in ovarian cancer.

Invasion and metastasis require the destruction of the extracellular matrix and basement membranes to facilitate growth or migration of tumor cells into vascular and lymphatic spaces. These processes are mediated by proteolytic enzymes. Malignant cells produce urokinase which is a protease known to enhance the invasiveness of many tumor cells. The relationship between urokinase and various prognostic factors was investigated in 16 patients with epithelial ovarian cancer. Tissue concentrations of urokinase were measured in tumor cytosols using enzyme-linked immunoassays. Urokinase levels were lower in ovarian tumors of low malignant potential (median 9.5, range 3.5-18.3 pg/mg protein, n = 4) than invasive cancers (median 44.6, range 16.1-210.6 pg/mg protein, n = 12), p < 0.01. In the invasive carcinomas urokinase levels did not vary significantly with tumor stage or cell type. Grade 3 tumors had higher levels of urokinase (median 120.4, range 21.4-397.1 pg/mg protein, n = 6) than grade 1 and 2 tumors (median 29.2, range 16.1-51.8 pg/mg protein, n = 6), p < 0.05. Urokinase levels were higher in recurrent (median 120.4, range 51.8-210.6 pg/mg protein, n = 4) than in primary (median 29.2, range 16.1-97.1 pg/mg protein, n = 8) tumors, p < 0.05. These results support the hypothesis that urokinase plays a role in invasion and metastasis of ovarian epithelial cancers and suggest that tissue levels of urokinase may have prognostic value.