The role of ultrasound in patients with possible penetrating cardiac wounds: a prospective multicenter study

BACKGROUND

Ultrasound is quickly becoming part of the trauma surgeon's practice, but its role in the patient with a penetrating truncal injury is not well defined. The purpose of this study was to evaluate the accuracy of emergency ultrasound as it was introduced into five Level I trauma centers for the diagnosis of acute hemopericardium.

METHODS

Surgeons or cardiologists (four centers) and technicians (one center) performed pericardial ultrasound examinations on patients with penetrating truncal wounds. By protocol, patients with positive examinations underwent immediate operation. Vital signs, base deficit, time from examination to operation, operative findings, treatment, and outcome were recorded.

RESULTS

Pericardial ultrasound examinations were performed in 261 patients. There were 225 (86.2%) true-negative, 29 (11.1%) true-positive, 0 false-negative, and 7 (2.7%) false-positive examinations, resulting in sensitivity of 100%, specificity of 96.9%, and accuracy of 97.3%. The mean time from ultrasound to operation was 12.1+/-5 minutes.

CONCLUSION

Ultrasound should be the initial modality for the evaluation of patients with penetrating precordial wounds because it is accurate and rapid.

Accelerated radiotherapy regimen for malignant gliomas using stereotactic concomitant boosts for dose escalation

The purpose of this pilot study was to determine the feasibility and toxicities of an accelerated treatment program by using a concomitant stereotactic radiotherapy boost given weekly during a course of standard external-beam irradiation (EBXRT) in patients with malignant gliomas. Twelve patients underwent biopsy or subtotal resection of a malignant glioma and were enrolled on the protocol, which delivered 44 Gy-EBXRT and a 12-Gy stereotactic radiotherapy boost given on 3 consecutive weeks of treatment for a total dose of 80 Gy over 33 days. Three patients with anaplastic astrocytoma and nine patients with glioblastoma multiforme had median survival times of 33 months and 16 months, respectively. All of the tumor recurrences were within or were closely adjacent to the region of high-dose irradiation. None of the patients required a treatment break, and there were no acute complications. Two patients developed seizures in the follow-up period, and four patients were diagnosed with radionecrosis at the time of the second operation. The treatment program was found to be feasible and was well tolerated, and it resulted in a rate of late complications similar to those of radiosurgery or interstitial brachytherapy.

The clinical complexity of splenic vein thrombosis

Upper gastrointestinal hemorrhage secondary to splenic vein thrombosis is a curable form of localized portal hypertension when treated with splenectomy. A high index of suspicion is necessary in order to promptly diagnose and treat this underrecognized condition that is most commonly caused by inflammation or neoplasm of the pancreas. The triad of isolated gastric varices, splenomegaly, and normal hepatic function is classic; it is not uncommon, however, for patients to have only some or even none of these conditions. Mesenteric angiography with venous phase imaging is the gold standard of diagnosis. Ultrasound and CT may identify splenic vein thrombosis, but are most helpful in delineating concomitant upper abdominal pathology. Early recognition and intervention allow associated underlying conditions to be treated under the same anesthetic with minimal morbidity and mortality.

Glycated albumin stimulates fibronectin gene expression in glomerular mesangial cells: involvement of the transforming growth factor-beta system

Albumin modified by Amadori glucose adducts, formed in increased amounts in diabetes, stimulates collagen IV production and gene expression in renal glomerular mesangial cells, and induces mesangial matrix accumulation accompanied by increased mRNA encoding alpha 1 (IV) collagen and fibronectin in diabetic animals. These effects contribute to the pathogenesis of diabetic nephropathy, and resemble biologic activities of the cytokine TGF-beta 1, which also has been causally implicated in diabetic renal disease. We postulated that Amadori-modified glycated albumin modulates TGF-beta 1 expression in mesangial cells, and that TGF-beta 1 participates in mediating the glycated albumin-induced increases in mesangial cell matrix production. To test this hypothesis, we measured mRNA encoding TGF-beta 1, the TGF-beta Type II receptor and fibronectin, a key matrix component of the TGF-beta 1 tissue response, after incubation of mesangial cells with glycated albumin. Steady state levels of the mRNAs encoding for these proteins were stimulated when mesangial cells were cultured in the presence of albumin containing Amadori glucose adducts compared with levels in cells cultured with the nonglycated, glucose-free counterpart. The glycated protein-induced changes in mRNA expression were observed with concentrations of glycated albumin encompassing those found in clinical specimens and in media containing physiologic (5.5 mM) glucose concentrations, indicating that they were due to the glucose-modified protein and not to a hyperglycemic milieu. Further, they were accompanied by increased translated fibronectin protein, which was prevented with TGF-beta neutralizing antibody, as was the glycated albumin-induced increase in fibronectin mRNA. The findings indicate that Amadori-modified glycated albumin stimulates mesangial cell TGF-beta 1 gene expression by mechanisms that are operative under normoglycemic conditions. These data provide the first link between elevated glycated serum albumin concentrations and increased TGF-beta 1 bioactivity in the pathogenesis of mesangial matrix accumulation in diabetes.

Identification of p130Cas as a mediator of focal adhesion kinase-promoted cell migration

Previously we have demonstrated that focal adhesion kinase (FAK)-promoted migration on fibronectin (FN) by its overexpression in CHO cells is dependent on FAK autophosphorylation at Y397 and subsequent binding of Src to this site. In this report, we have examined the role of FAK association with Grb2 and p130(Cas), two downstream events of the FAK/Src complex that could mediate integrin-stimulated activation of extracellular signal-regulated kinases (Erks). We show that a Y925F FAK mutant was able to promote cell migration as efficiently as FAK and that the transfected FAK demonstrated no detectable association with Grb2 in CHO cells. In contrast, cells expressing a FAK P712/715A mutant demonstrated a level of migration comparable to that of control cells. This mutation did not affect FAK kinase activity, autophosphorylation, or Src association but did significantly reduce p130(Cas) association with FAK. Furthermore, FAK expression in CHO cells increased tyrosine phosphorylation of p130(Cas) and its subsequent binding to several SH2 domains, which depended on both the p130(Cas) binding site and the Src binding site. However, we did not detect increased activation of Erks in cells expressing FAK, and the MEK inhibitor PD98059 did not decrease FAK-promoted cell migration. Finally, we show that coexpression of p130(Cas) further increased cell migration on FN and coexpression of the p130(Cas) SH3 domain alone functioned as a dominant negative mutant and decreased cell migration. Together, these results demonstrate that p130(Cas), but not Grb2, is a mediator of FAK-promoted cell migration and suggest that FAK/ p130(Cas) complex targets downstream pathways other than Erks in mediating FAK-promoted cell migration.

Development of a custom designed TAH using rapid prototyping

The failure of the orthotopic implantation of a totally implantable artificial heart (TAH) was due mainly to anatomic mismatches in the conduits of the conventional TAH system. To overcome this anatomic incompatibility, a custom design and fabrication process was designed using the rapid prototyping (RP) technique. After three dimensional reconstruction of magnetic resonance imaging of the thoracic cavity and vascular remnants of the recipient, study of anatomic fit was done using the reconstructed thoracic model and three dimensional computer aided design (CAD) model. The direction of the inflow and outflow conduits of the blood sac was changed with a Unigraphics CAD. The RP model of the designed chamber was fabricated and examined for anatomic compatibility. Through this approach, the minute directional mismatch of the inflow and outflow conduits was improved. Thus, a new custom designed moving actuator Korean TAH with CAD and RP techniques was developed.

Ultrasound training during ATLS: an early start for surgical interns

OBJECTIVES

Determine if: (1) surgical interns could learn essential ultrasound principles of the focused assessment for the sonographic examination of the trauma patient (FAST) during Advanced Trauma Life Support (ATLS), and (2) swine are adequate models for learning the FAST.

DESIGN

Lecture, videotape, pre/posttests, and practical examination.

METHODOLOGY

Day 1: Survey, pretest, lecture, and videotape. Day 2: Three swine, used in ATLS, had diagnostic peritoneal lavage catheters reinserted to infuse fluid and produce "positive" ultrasound examinations. Two fresh swine were "negatives"; however, all five swine were draped similarly to disguise interventions. Interns were tested individually by surgeon-sonographers to determine whether the ultrasound image was "positive" or "negative." Posttests were completed while surgeons performed postmortem examinations on two swine.

STATISTICS

Paired Student's t test and Wilcoxon Rank Sum test.

RESULTS

Survey (5 min): 48% had exposure to ultrasound. Relationship of test scores showed no significant difference (p = 0.46 to 0.91) between interns with and without ultrasound experience. Tests (30 min): Mean pre- and posttests scores = 65.6 and 90.8, respectively (p < 0.001). Practical examination (140 min): Mean score = 89.6. Postmortem examinations: Left hepatic lobe partially obscured the spleen.

CONCLUSIONS

(1) Surgical interns can learn essential ultrasound principles of the FAST during ATLS (2) Swine are feasible models for learning the FAST.

In situ detection of hepatitis C virus RNA in liver tissue using a digoxigenin-labeled probe created during a polymerase chain reaction

The cellular localization of hepatitis C virus (HCV) RNA in liver tissue was studied by nonisotopic in situ hybridization using a digoxigenin-labeled cDNA probe created during a polymerase chain reaction on samples from 16 patients with chronic HCV infection. Hybridization signals were recognized in the cytoplasm of the hepatocytes, and a few hepatocytes had hybridization signals in the nucleus as well. HCV RNA positive hepatocytes were found in 1 of 9 patients with chronic persistent hepatitis, 2 of 5 patients with chronic active hepatitis, and in each of 2 patients with chronic active hepatitis and cirrhosis. Positive signals were found in many hepatocytes within the lobule in liver sections of patients with advanced chronic active hepatitis. A number of HCV RNA positive hepatocytes were found in nodules, but not in the area of fibrosis. On the other hand, positive signals were found in a few hepatocytes scattered in the lobule in a patient with chronic persistent hepatitis. The mean ALT levels in the patients with positive signal (175.6 +/- 44.2 U/L) were significantly higher than in those without a signal (70.27 +/- 16.1 U/L) (P < 0.05). The findings suggest that a larger amount of HCV may be present during the advanced than during the early stages of type C hepatitis and nonisotopic in situ hybridization using a digoxigenin-labeled HCV cDNA probe created during a polymerase chain reaction deserves wider application for the detection of HCV replication in specimens.

Surgical treatment of acute myocardial infarct: clinical considerations

Most clinicians have become convinced of the fundamental role of an early reperfusion strategy in the improved outcome for patients suffering acute myocardial infarction. An increasing body of work suggests that the surgeon has the best opportunity for control of the conditions of acute reperfusion, attenuation of secondary injury, and provision of superior myocardial salvage compared with thrombolytic or angioplasty therapy. Before general acceptance or implementation of surgical treatment as an initial treatment for certain acute infarct patient subgroups, the policy must be compared with our current standards of care in a prospective, randomized fashion. The available data suggest that such inclusion of a primary surgical option in trials investigating treatment of acute myocardial infarct is long overdue.

Host recognition by the VirA, VirG two-component regulatory proteins of agrobacterium tumefaciens

Agrobacterium tumefaciens contains about 25 vir genes localized on a 200-kb tumour-inducing (Ti) plasmid that direct a conjugation-like transfer of tumorigenic DNA from the bacterium to the nuclei of infected plant cells. These genes are strongly and coordinately induced during infection in response to three different classes of stimuli which are thought to be key chemical features of a typical wound site. These stimuli are (i) guaiacol and syringol derivatives such as acetosyringone, (ii) sugars such as glucose and glucuronic acid, and (iii) acidic pH. The sensing of these compounds is carried out by the VirA, VirG and ChvE proteins. VirA is a four-domain histidine protein kinase, while VirG is a transcriptional activator which is activated by VirA-mediated phosphorylation. ChvE is a chromosomally encoded periplasmic sugar binding protein which is required for sensing sugars but dispensable for sensing the other two stimuli. Here we will review the nature of these chemical stimuli, the structure and function of the three regulatory proteins, their similarity to sensors found in human and animal pathogens, the factors influencing their pool size, and their role in the host range of different strains of A. tumefaciens.

A mutation in the receiver domain of the Agrobacterium tumefaciens transcriptional regulator VirG increases its affinity for operator DNA

We fused the wild-type Agrobacterium tumefaciens virG gene and the constitutive virGN54D allele to the malE gene of Escherichia coli, and studied the binding of MBP-VirG fusions to the autoregulated virG promoter. MBP-VirGN54D protein bound this promoter with 10-fold higher affinity than MBP-VirG, and bound to vir box I with eightfold higher affinity than to vir box III. Disruption of vir box III did not alter the affinity for vir box I, suggesting a lack of cooperativity between these sites. We provide evidence that protein bound at a single vir box may have a higher oligomeric state than non-bound protein, and that a DNA distortion adjacent to vir box I may occur during activation.

Altered-function mutations of the transcriptional regulatory gene virG of Agrobacterium tumefaciens

Three point mutations were isolated in the Agrobacterium tumefaciens virG gene by screening for vir gene expression in the absence of added phenolic inducing compounds. All three mutations were localized in the predicted amino-terminal phosphoryl receiver domain of the protein. One mutant (N54D) bypasses the requirement for VirA and phenolic inducers both for transcriptional activation of all tested vir promoters and for plant tumorigenesis. This mutant also activates vir gene expression efficiently at neutral pH, indicating that the step in induction that is normally stimulated by acid pH occurs before or during VirG phosphorylation. The other two mutants (M13T and H15R) require VirA for activity but are sensitized to low levels of inducing stimuli.

Effects of erythropoietin-induced hemopoiesis on peritoneal transport and on in vitro T cell response in CAPD

In order to investigate the therapeutic efficacy of subcutaneously administered erythropoietin (rHuEPO) and the effects of rHuEPO-induced hemopoiesis on peritoneal transport and on cellular immune responses, we performed standardized peritoneal equilibration tests and measured T cell subsets and phytohemagglutinin (PHA)-induced interleukin-2 receptor (IL-2R) expression of PBMC by flow cytometry before and after subcutaneous rHuEPO (Eprex-, Cilag), 4000 U twice weekly, in 13 stable CAPD patients. Hct increased from 21.3 +/- 3.4% to 30.0 +/- 4.8% after 1 mo and to 32.7 +/- 4.9% after 2 mon of rHuEPO. Drained volume after 4 hrs of dwell with 4.25% dialysate increased from 2,675 +/- 204 ml to 2,807 +/- 174 ml (P < 0.05). D4/P4 creatinine increased from 0.68 +/- 0.07 to 0.71 +/- 0.06 (P < 0.05) and creatinine clearance from 7.57 +/- 0.71 to 8.03 +/- 0.63 ml/min (P < 0.05). The number of total circulating lymphocytes, T4,T8, T4/T8 with or without PHA did not change after rHuEPO. PHA-induced IL-2R expression by PBMC as expressed by mean channel of fluorescence intensity increased from 149.8 +/- 6.7 to 156.8 +/- 6.1 (P < 0.05).

CONCLUSION

Subcutaneous rHuEPO is effective in correcting anemia in CAPD patients. rHuEPO-induced hemopoiesis is associated with increase in peritoneal creatinine and water transport and also with PHA-induced IL-2R expression.

Subcutaneously implanted catheters reduce the incidence of peritonitis during CAPD by eliminating infection by periluminal route

Recent experiences with Y-connectors suggest that the flush-before-fill effectively reduces intraluminal infection. Periluminal infection, however, remains an important route of peritonitis (P). We have recently reported reduced P incidence with the introduction of a new access technique as described by Moncrief in which the external segment of peritoneal catheter is left implanted subcutaneously for 6 weeks before exteriorization and bag exchanges. P developed once every 14.0 patient-mos with the new access while the incidence was one episode per 10.7 mos with conventional access. Significantly fewer patients with the new access compared to those with conventional access experienced P during the observation period (p < 0.01). Although the overall incidence of exit-site infection (ESI) was not different, there were significantly fewer episodes of simultaneous P and ESI with the new access (2P in 47 episodes of ESI) than with conventional access (36P in 126 ESI). While 10 of the 36 episodes of simultaneous infection in the conventional technique were caused by same organisms, none of the 2 episodes with the new access technique was caused by same organisms.

CONCLUSION

The results of this study suggest that the new access technique reduces P incidence by virtually eliminating infection by the periluminal route.

Metabolic correlates of adenosine formation in stimulated guinea pig heart

Adenosine release into epicardial fluid and coronary effluent of isolated isovolumic guinea pig hearts was examined at baseline and after stimulation with norepinephrine (30 nM) during 31P-nuclear magnetic resonance spectroscopy to monitor myocardial metabolism. At baseline flow (9.6 +/- 0.3 ml.min-1.g-1), epicardial and venous adenosine concentrations were 154 +/- 40 and 17 +/- 5 nM, respectively. The phosphorylation potential (log[ATP]/[ADP][Pi]) and the phosphocreatine-inorganic phosphate ratio ([PCr]/[Pi]) were 5.26 +/- 0.04 and 8.5 +/- 0.7, respectively. Norepinephrine increased left ventricular pressure, heart rate, and myocardial O2 consumption rate by approximately 21, 70, and 45%, respectively, and increased epicardial and venous adenosine to 496 +/- 74 and 461 +/- 94 nM, respectively. Log-[ATP]/[ADP][Pi] and [PCr]/[Pi] declined to 4.57 +/- 0.06 and 1.9 +/- 0.3, respectively. Epicardial [AMP] increased from 54 +/- 13 to 123 +/- 24 nM. AMP was not detectable in the venous effluent. Coronary resistance correlated with epicardial and venous [adenosine] (r = 0.86 and 0.90). Epicardial and venous [adenosine] correlated with log[ATP]/[ADP][Pi], [PCr]/[Pi], and cytosolic [AMP]. Hence, interstitial adenosine is linked to cytosolic metabolism and may regulate coronary vascular resistance. Venous adenosine underestimates epicardial adenosine at baseline but more closely approximates epicardial adenosine during norepinephrine infusion.

Inhibition of 125I organification and thyroid hormone release by interleukin-1, tumor necrosis factor-alpha, and interferon-gamma in human thyrocytes in suspension culture

To elucidate the mechanism of decreased 131I uptake by the thyroid gland in patients with subacute thyroiditis and painless thyroiditis, human thyroid follicles were cultured with interleukin-1 (IL-1), tumor necrosis factor-alpha (TNF alpha), and/or interferon-gamma (IFN gamma), and the effects of these cytokines on thyroid function were studied in vitro. When human thyrocytes were cultured in RPMI-1640 medium containing 0.5% fetal calf serum and TSH for 5-8 days, the cells incorporated 125I, synthesized de novo [125I]iodotyrosines and [125I]iodothyronines, and secreted [125I]T4 and [125I]T3 into the medium. IL-1 alpha and IL-1 beta inhibited 125I incorporation and [125I]iodothyronine release in a concentration-dependent manner. The minimal inhibitory effect was detected at 10 pg/ml. Electron microscopic examination revealed a marked decrease in lysosome formation in IL-1-treated thyrocytes. TNF alpha and IFN gamma also inhibited thyroid function in a concentration-dependent manner. Furthermore, when thyrocytes were cultured with IL-1, TNF alpha and IFN gamma, these cytokines more than additively inhibited thyroid function. Although the main mechanism of 131I uptake suppression in the thyroid gland in subacute thyroiditis is due to cellular damage and suppression of TSH release, our present findings suggest that IL-1, TNF alpha, and IFN gamma produced in the inflammatory process within the thyroid gland further inhibit iodine incorporation and at least partly account for the decreased 131I uptake by the thyroid gland in destruction-induced hyperthyroidism.

3,3',5'-Triiodothyronine inhibits ontogenetic development of iodothyronine-5'-deiodinase in the liver of the neonatal mouse

To elucidate the effect of rT3 on iodothyronine-5'-deiodinating activity (I-5'-DA) in the liver of neonatal mice, rT3 was injected sc on the 5-8th day after birth and I-5'-DA in the liver was determined. A single injection of rT3 (0.01-1 microgram/g) inhibited the ontogenetically developing I-5'-DA in a dose- and time-dependent manner. The inhibitory effect was reversible and specific for I-5'-DA. Lineweaver-Burk analysis revealed that the time- and dose-dependent decrease in the enzyme activity was due to a decrease in Vmax with no alteration in Km values (5 x 10(-8) mol/l). The maximal inhibitory effect was observed at a dose of 1 microgram rT3/g, whereas the inhibitory effect was diminished at greater doses (4-10 micrograms/g), probably owing to a contamination with T4 of the rT3 preparation administered. Furthermore, consistent with our previous in vitro findings, rT3 inhibited the I-5'-DA induced by T3 in the liver of neonatal mice. These findings suggest that rT3 inhibited I-5'-DA in the liver of neonatal mice by decreasing the amount of enzyme available to the substrate and that rT3 also elicited an antagonistic effect against T3 in the induction of I-5'-DA in vivo.

Effects of tumor necrosis factor-alpha/cachectin on thyroid hormone metabolism in mice

To elucidate the mechanism by which low T3 and low T4 syndrome occurs in patients with acute or chronic infection or malignancy, recombinant human tumor necrosis factor-alpha (TNF-alpha)/cachectin (TNF) was administered ip to mice and thyroid hormone metabolism was studied. Since administration of TNF caused a decrease in food intake and body weight, all experiments were performed using pair-fed control (PFC) mice. Administration of TNF at a dose of 1-100 micrograms/day for 3 days decreased serum T4, T3, and rT3 concentrations in a dose-dependent manner. In PFC mice, serum T4 and T3 also decreased, but rT3 was significantly increased. T3/T4 ratio was greater in TNF-treated mice than in PFC mice. Type I iodothyronine-5'-deiodinating activity in the liver was significantly decreased in PFC mice but not in TNF-treated mice. The effect of TNF was reversible and could be abolished by boiling the cytokine. Furthermore, T3 and T4 response to TSH was greatly diminished in TNF-treated mice in comparison with PFC mice. These findings suggest that TNF directly inhibited the effect of TSH on the thyroid gland and decreased the serum concentrations of T4 and T3. Although TNF decreased food intake and body weight in TNF-treated mice, it did not decrease type I 5'-deiodinating activity in the liver, resulting in a greater T3/T4 ratio and lower serum rT3 concentration than those in PFC mice. We speculate that TNF is at least partly involved in the altered thyroid hormone metabolism (decreased serum T4, T3, and rT3 concentrations) caused by infections in mice.

Stimulation of alkaline phosphatase activity by thyroid hormone in mouse osteoblast-like cells (MC3T3-E1): a possible mechanism of hyperalkaline phosphatasia in hyperthyroidism

In order to elucidate the mechanism of increased alkaline phosphatase (AI-P) activity of bone origin in serum of patients with hyperthyroidism, the effects of thyroid hormone on mouse osteoblast-like cells (MC3T3-E1) were studied in vitro. Triiodo-L-thyronine (T3) and thyroxine (T4) produced a dose-dependent increase in AI-P activity in the cells at minimum concentrations of 10(-10)M T3 (free T3, 5 x 10(-12) M) and 10(-8) M T4 (free T4, 8 x 10(-11) M), respectively. Scatchard analysis revealed that MC3T3-E1 cells contained nuclear binding sites specific for T3 with an apparent Kd of 120 pM (maximum number of binding sites, approximately 2500 per cell). When cells were cultured with T3 in alpha-minimal essential medium (alpha-MEM) for a prolonged period, AI-P activity also became detectable in the conditioned medium. In contrast to rat osteosarcoma cells (ROS 17/2.8), MC3T3-E1 cell growth was inhibited by T4 in a concentration-dependent manner. These findings suggest that thyroid hormone inhibits proliferation and stimulates differentiation of mouse osteoblast-like cells. Since T3 and T4 stimulate AI-P activity not only in the cells but also in the medium, we speculate that the hyper-alkaline phosphatasia frequently seen in patients with hyperthyroid Graves' disease is partly due to a direct effect of thyroid hormone on osteoblasts or osteoblast-like cells.

Dimethylsulfoxide maintains human thyroid cells in suspension culture, facilitating synthesis and release of thyroid hormone

Usually, human thyrocytes in primary culture rapidly lose their thyroid function and fail to synthesize or release thyroid hormone after 3-5 days of culture. By culturing thyroid follicles obtained from patients with Graves' disease in medium supplemented with TSH and a low concentration of fetal calf serum (1%), thyrocytes can maintain thyroid function for several days. We have found that the addition of dimethylsulfoxide to culture medium (1.7%) furthermore enhanced and maintained thyroid function (de novo synthesis and release of [125I] thyroxine) for more than 13 days, probably by inhibiting dedifferentiation of thyrocytes. The present bioassay will be also useful for detecting thyroid stimulating immunoglobulin in patients with Graves' disease.

A highly sensitive bioassay for PTH using ROS 17/2.8 subclonal cells

A highly sensitive bioassay for PTH was developed by using rat osteosarcoma cells (ROS 17/2.8). By limiting dilution, ROS cells were subcloned and the subclonal cell line (ROS 17/2.8-5) most responsive to PTH was selected. When subconfluent ROS 17/2.8-5 cells were treated with hydrocortisone for 3 days and then incubated with PTH, the cAMP response was significant at 10-40 ng/l hPTH (1-34) (4 approximately 16 X 10(-12) mol/l). Osteoclast activating factors such as human interleukin 1 alpha and beta, and tumour necrosis factor alpha did not stimulate cAMP production, whereas a conditioned medium of oesophageal carcinoma cells established from a patient with humoral hypercalcaemia stimulated cAMP production. By selecting PTH-responsive subclonal cells and treating them with hydrocortisone, the sensitivity for detecting PTH was improved approximately 15 times. This method will be useful in the characterization and purification of PTH-like factors produced by malignant tumours from hypercalcaemic patients.