Copeptin as a non-invasive biomarker in chronic thromboembolic pulmonary hypertension

Introduction

Copeptin is the C-terminal fragment of the precursor protein of vasopressin. In acute pulmonary embolism, copeptin has been suggested to be a strong predictor of outcome and to provide additional predictive value to the established cardiac biomarkers high-sensitivity cardiac troponin and N-terminal pro-brain natriuretic peptide (NT-proBNP). Chronic thromboembolic pulmonary hypertension (CTEPH) is diagnosed in about 5% of patients who survive acute pulmonary embolism. Individualized risk stratification remains a challenge in the work-up of CTEPH patients.

Purpose

The current study investigated whether copeptin has the potential to aid the stratification of patients who have experienced pulmonary embolism and CTEPH patients. We examined the baseline (BL) levels and dynamics of copeptin during therapy in CTEPH patients who underwent balloon pulmonary angioplasty (BPA) or pulmonary endarterectomy (PEA). Moreover, the study compared copeptin levels between patients with or without therapy response.

Methods

The study included a total of 125 CTEPH patients scheduled for treatment. A total of 78 underwent staged BPA and 64 underwent PEA. In accordance with recent studies from our group, therapy success was defined as a decrease in meanPAP ≥25% and PVR ≥35% or a normalization below the thresholds defining pulmonary hypertension. Blood samples were collected at BL, prior to each BPA session in the BPA cohort, and at follow-up (FU) 6 months after BPA or 12 months after PEA. Copeptin was measured in thawed serum aliquots by an immunochemical method.

Results

The 78 patients in the BPA cohort underwent a mean of 6 BPA procedures each; there were a total of 413 interventions. The hemodynamic clinical and functional status the CTEPH patients improved after BPA and PEA therapy: meanPAP (BL: 43±9 mmHg vs. FU: 27±9 mmHg; p<0.001); PVR (BL: 7.6±3.4 WU vs. FU: 3.8±2.0 WU; p<0.001); RAP (BL: 7.9±5.8 mmHg vs. FU: 5.4±2.7 mmHg; p<0.001); WHO functional class [BL: I:0 / II:25 / III:80 / IV:20 vs. FU: I:56 / II:57 / III:10 / IV:2]; 6-minute-walk distance (BL: 405±99 m vs. FU: 456±112 m; p<0.001).

The median serum levels of copeptin [BL 7.7 (4.6–14.2) pmol/L vs. FU 6.3 (3.9–12.5); p=0.009] and NT-proBNP [BL: 811 (157–1857) ng/L vs. FU: 142 (72–335) ng/L p<0.001] decreased significantly after therapy. The copeptin levels did not correlate with hemodynamics at BL: PVR (rrs=0.02; p=0.79) and meanPAP (rrs=0.03; p=0.75). The copeptin levels at BL (AUC=0.61) and the relative change (AUC=0.53) did not predict the endpoint of therapy response.

Conclusions

Copeptin levels are elevated in CTEPH patients compared with normal values in the literature. Although copeptin is known to provide additional value in the context of risk stratification in acute pulmonary embolism, it failed to provide additional diagnostic benefit in CTEPH in the current study.

Funding Acknowledgement

Type of funding source: Public grant(s) – National budget only. Main funding source(s): SFB 1213 area CP01

Small Scale Intramolecular Flexibility in 111mCd-Plastocyanin

The effect of mutations in the vicinity of the putative electron transfer path on the metal center of the electron transfer protein plastocyanin (spinacea) is investigated by monitoring the nuclear quadrupole interaction of 111mCd in Cd-derivatives of the protein via time differential perturbed angular correlation. The spectra for the wild type protein and the mutants were rather similar. All spectra exhibit a peculiar line profile which points towards a small scale intramolecular flexibility of the metal center.

Hg-Coordination Studies of Cysteine Containing Oligopeptides

In order to study the interaction of cysteine containing peptide chains with Hg(II) the nuclear quadrupole interaction (NQI) of 199mHg in the Hg complex of the oligopeptide Alanine-Alanine-Cysteine-Alanine-Alanine (AACAA) was determined by time differential perturbed angular correla-tion. Different 199mHg-NQI’s for free and resin-bound AACAA were obtained. Furthermore, the 199mHg-NQI’s are influenced by the Hg:AACAA stoichiometry.

[Gemcitabine/cisplatin vs. MVAC. 5 year survival outcome of the phase III study of chemotherapy of advanced urothelial carcinoma in Germany]

Of 405 patients with stage IV transitional cell carcinoma from an international multicenter phase III trial, 70 were randomized in Germany to receive either gemcitabine/cisplatin or standard MVAC systemic chemotherapy for locally advanced or metastatic urothelial cancer. Overall survival as the primary endpoint of the study was similar in both arms (median survival GC 15.4 months vs MVAC 16.1 months), as were tumor-specific survival and time to progressive disease. In the intent-to-treat analysis, the 5-year overall survival rate was 10% for patients randomized to GC and 18% randomized to MVAC. Tumor overall response rates (GC 54%, MVAC 53%) were similar. The toxic death rate was 0% in the GC arm and 3% (one patient) in the MVAC arm. Significantly more GC than MVAC patients experienced grade 3/4 anemia (GC 52%, MVAC 20%) with significantly more red blood cell transfusions in the GC arm.Significantly more GC than MVAC patients had grade 3/4 thrombocytopenia (GC 54%, MVAC 17%) without grade 3/4 hemorrhage or hematuria in either arm. More MVAC patients experienced grade 3/4 neutropenia (GC 56%, MVAC 61%, p=1.000), neutropenic or leukopenic fever (GC 0%, MVAC 10%, p=0.237), mucositis (GC 0%, MVAC 7%, p=0.495), and alopecia (GC 6%, MVAC 36%, p=0.004). GC represents a reasonable alternative for the palliative treatment of patients with locally advanced and metastatic transitional cell carcinoma. Sustained long-term survival was only found for patients with locally advanced cancer, lymphatic metastases, or solitary distant metastasis but not for visceral metastatic disease.

Tibolone versus 17beta-estradiol/norethisterone: effects on the proliferation of human breast cancer cells

Tibolone is a synthetic progestin with estrogenic and progestogenic properties, widely used for alleviation of menopausal syndromes and for osteoporosis prophylaxis in postmenopausal women. Since only little data are available on tibolone and breast cancer risk the present study investigates the effect of tibolone on the growth of the human breast cancer cell line, MCF-7. Tibolone is clinically comparable to an estradiol/norethisterone combination, therefore we included this hormone combination in our experiments. Tibolone was examined alone and in the presence of 0.1 nM estradiol in the concentration range from 0.001 microM to 1 microM. Norethisterone was studied using the same concentration range in combination with 0.1 nM estradiol. Tibolone led to significant cell growth in the concentration range of 0.01 to 1 microM and was able to significantly stimulate estradiol-induced proliferation at the concentrations 0.01 and 0.1 microM. In contrast, the estradiol/norethisterone combination elicited significant inhibition of cell growth at the concentrations 0.001 and 0.01 microM. These data suggest that tibolone does have tumor cell-growth promoting effects in vitro whereas the estradiol/norethisterone combination partially inhibits cell growth. Therefore no differences in risk profile are to be expected between conventional hormone substitution using estradiol and norethisterone acetate and tibolone. Drawing a clinical consequence from our experiments would result in not recommending the use of tibolone in postmenopausal women at high risk for breast cancer development until long-term controlled clinical studies have been performed on the effect of tibolone administration and breast cancer risk.

Comparison of the effects of 17alpha-ethinylestradiol and 17beta-estradiol on the proliferation of human breast cancer cells and human umbilical vein endothelial cells

OBJECTIVE

Little is known about the direct comparison between 17alpha-ethinylestradiol (EE2) and 17beta-estradiol (E2) on pre-existing breast cancer cells and on their angiogenetic effects. In this study we investigated the effect of both estrogens on the proliferation of MCF-7 cells, a human breast cancer cell model, and on human umbilical vascular endothelial cells (HUVECs).

METHODS

The steroids were tested in the concentration range of 10(-10) to 10(-5) M. The proliferation of MCF-7 and HUVEC cells was measured after five and six days by the crystal violet staining technique.

RESULTS

In the concentration range from 10(-10) to 10(-5) M both estrogens showed a proliferative effect in the MCF-7 cells, E2 having a gradually declining effect on proliferation with increasing concentration while EE2 showed a constant proliferative effect over a large concentration range. At the highest concentration tested E2 had no effect on proliferation while EE2 even inhibited growth. In the HUVEC cells both estrogens showed a slightly significant stimulatory effect at the lowest concentration, and no significant effect at the remaining concentrations. EE2 again inhibited cell growth at the highest concentration.

CONCLUSIONS

At the serum concentrations seen in hormone replacement therapy, EE2 appears to have less proliferative effect on breast cancer cells compared with E2, while both estrogens appear to have similar effects on endothelial cells.

The effect of endogenous estradiol metabolites on the proliferation of human breast cancer cells

Evidence is accumulating that estradiol metabolites may be involved in carcinogenesis as some metabolites exert proliferative and others anti-proliferative properties on human cancer cells. The present study is the first to investigate the effect of 14 endogenous estradiol metabolites on the proliferation of the human breast cancer cell line, MCF-7, in comparison with the effect of the parent substance 17beta-estradiol with special concern on high pharmacological concentrations. The steroids were tested in the range from 10(-8) to 10(-5) M on MCF-7 cells which were incubated for nine days. Estradiol and almost all A-ring metabolites displayed biphasic reactions on cell proliferation, i.e. stimulatory at low concentrations and inhibitory at the highest concentration, 10(-5) M. The D-ring metabolites did not show such clear biphasic patterns, in most of them the stimulatory effect prevailed at the highest dosage used. The strongest inhibitory effect was seen for the A-ring metabolite 2-methoxyestradiol at the concentrations of 10(-6) and 10(-5) M and the strongest stimulatory effect was noted for the D-ring metabolite estriol at the same concentrations. The results indicate that some A-ring metabolites might be suitable for breast cancer treatment when used in high dosages. This is of special interest, since many of these metabolites have very weak estrogenic activity.

Effects of tibolone on human breast cancer cells and human vascular coronary cells

Tibolone is a synthetic progestin with estrogenic and progestogenic properties, used to alleviate menopausal syndromes and for osteoporosis prophylaxis in postmenopausal women. However, only little data are available on tibolone and breast cancer risk and on the effects of tibolone on the cardiovascular system. Therefore, in the present in vitro study, we investigated the effect of tibolone on the growth of the human breast cancer cell line, MCF-7, and on the direct effects of tibolone on the vasculature, i.e. human female coronary endothelial and smooth muscle cells. In the breast cancer cell experiments, tibolone was examined alone and in the presence of 0.1 nM estradiol in the concentration range from 0.001 microM to 1 microM. Tibolone lead to significant cell growth in the concentration range of 0.01 micro M to 1 microM and was not able to inhibit estradiol-induced proliferation at the concentrations of 0.01 microM and 0.1 microM. In the vascular endothelial cell culture experiments, tibolone was tested at the concentrations 0.1 microM, 1 microM and 10 microM. Tibolone reduced the synthesis of endothelin as well as the concentrations of E-selectin, PAI-1 and pro-MMP-1. The magnitude of the effects on these markers varied and was in the range of 11%-42%. Concerning smooth muscle cells, tibolone elicited no changes in the proliferation compared to control values. These data suggest that tibolone does have tumour cell-growth promoting effects in vitro. However, tibolone can positively influence the synthesis of markers in cell cultures of human female coronary artery, which modulate vascular tone and which play a decisive role in the various stages of atherosclerosis. Drawing a clinical consequence from our experiments would result in not recommending the use of tibolone in postmenopausal women at high risk for breast cancer development until long-term controlled clinical studies have been performed on the effect of tibolone administration and breast cancer risk. Experimental studies, such as the present one, are useful to explore mechanisms, but clearly cannot replace clinical studies.

[Gemcitabine in advanced bladder cancer]

MVAC (methotrexate, vinblastine, doxorubicin, cisplatin) has been the standard treatment for patients with advanced urothelial cancer for more than 15 years. Combination chemotherapy including gemcitabine/cisplatin showed similar tumor response and survival rates with a more tolerable toxicity profile in a recent multinational phase III study when compared to MVAC. Effectiveness of gemcitabine as a single agent or in combination with other cytotoxic agents had been investigated before in several phase II studies treating patients with advanced urothelial cancers. The tumor response rate for single agent gemcitabine in advanced urothelial cancers is between 11% and 28%. Tumor response rates rise to 50% when combining gemcitabine with cisplatin, and median survival times between 12 and 15 months can be expected. Triplet therapy schedules including gemcitabine may yield response rates in up to 80% of patients, particularly when used sequentially with other regimens. Further improvement of tolerability during systemic gemcitabine/cisplatin combination therapy without compromising effectiveness was recently demonstrated by a German phase II study when the 4-week schedule was reduced to a 3-week schedule with gemcitabine given on days 1 and 8.

The effect of medroxyprogesterone acetate and norethisterone on the estradiol stimulated proliferation in MCF-7 cells: comparison of continuous combined versus sequential combined estradiol/progestin treatment

OBJECTIVE

Little is known on the type of progestin and regimen type in relation to breast cancer risk. We have compared the effect of medroxyprogesterone acetate (MPA) and norethisterone (NET) on the estradiol stimulated proliferation in MCF-7 cells with respect to different regimens used in combined hormone replacement therapy (HRT).

DESIGN

To approximate the in vivo conditions in HRT, MCF-7 cultures were pretreated with estradiol followed by estradiol/progestin treatment to represent the sequential combined model and compared with non pretreated cultures followed by estradiol/progestin treatment for the continuous combined model.

RESULTS

When using progestins in the continuous combined form with estradiol (10(-10) M) both progestins showed a significant reduction in the estradiol stimulated proliferation of the MCF-7 cells. In the sequential combined model the addition of MPA led to a stronger significant reduction of MCF-7 proliferation but in a narrower concentration range (from 10(-8) to 10(-6) M) compared to the continuous treatment. NET did not show any significant effect on proliferation in the SC model.

CONCLUSION

Different regimen types and different progestins do lead to significantly different effects on the proliferation of a breast cancer cell line. These findings might be useful in the elucidation of potential mechanisms involved in the clinical situation.

Valsartan and candesartan can inhibit deteriorating effects of angiotensin II on coronary endothelial function

The angiotensin II (Ang II) AT1-receptor antagonists, valsartan and candesartan, were compared with regard to their effect on Ang II-mediated changes in parameters of coronary endothelial function. Ang II (10 microM) induced increased concentrations of the vasoconstrictor endothelin, the procoagulatory substance plasminogen-activator-inhibitor-1 (PAI-1) and the precursor of the matrix-metalloproteinase 1 (MMP-1) in endothelial cell cultures from human coronary arteries. These increases were completely prevented by the addition of 10 microM valsartan or candesartan and partially by the addition of lower concentrations of these drugs, i.e. 1 microM and 0.1 microM. No significant difference between the effect of the two AT1-receptor antagonists was observed. These results suggest that AT1-receptor antagonists not only can reduce blood pressure by blocking the action of Ang II, but might also contribute to the prevention of atherogenesis and plaque instability.

Comparison of the effects of continuous combined and sequential combined medroxyprogesterone acetate-estradiol treatment on the proliferation of MCF-7 cells

OBJECTIVE

The aim of the study was to examine the effects of two different, clinically relevant, hormone replacement therapy (HRT) regimen types, continuous combined and sequential combined, on breast cancer cells by means of an in vitro model. The study was carried out using the C21-progestin medroxyprogesterone acetate (MPA) in combination with estradiol, and with the proliferation of MCF-7, a human breast cancer cell-line, as end-point.

METHODS

Proliferation of MCF-7 cells was measured by means of a crystal violet staining technique. Growth was triggered using a constant estradiol concentration of 10(-10) mol/l, while varying the MPA concentration from 10(-11) to 10(-6) mol/l.

RESULTS

The continuous combined model of treatment led to the inhibition of estradiol-induced growth of MCF-7 with MPA concentrations of 10(-10) mol/l and upwards, compared with estradiol-alone-induced growth. The sequential combined model showed a greater inhibition at the higher MPA concentrations of 10(-8)-10(-6) mol/l, with reduced sensitivity to inhibition at the lower MPA concentrations tested, of 10(-11)-10(-9) mol/l. The different treatment types resulted in significantly different sensitivities of the MCF-7 cells to inhibition of estradiol-induced proliferation at the higher MPA concentrations of 10(-8)-10(-6) mol/l.

CONCLUSIONS

The results demonstrate the importance of considering in vivo factors in an in vitro model with regard to improving the cell culture techniques used, to obtain a clearer picture of the possible mechanisms involved in the potential breast cancer risk with different HRT regimens.

The effects of A-ring and D-ring metabolites of estradiol on the proliferation of vascular endothelial cells

The effects of 14 estradiol metabolites on the proliferation of cultured endothelial cells of human umbilical cord veins were examined and compared with that of their parent substance estradiol. The relationship between dosage and effect was tested over the pharmacological concentration range of 10(-8) to 10(-5) M. Estradiol showed a biphasic behaviour, in the form of stimulation at low concentrations and inhibition at the highest concentration. All 10 A-ring metabolites tested stimulated the growth of the endothelial cells at the lower concentrations. At the highest concentration, the 5 A-ring metabolites: 2-hydroxyestrone, 2-hydroxyestradiol, 2-hydroxyestriol, 4-hydroxyestrone and 4-hydroxyestradiol caused significant inhibitions. Except for the 2-hydroxyestradiol, methylation of these metabolites resulted in the loss of the proliferation inhibiting effect. The D-ring metabolites showed no marked effects compared to the A-ring metabolites except for 16alpha-hydroxyestrone which had an inhibiting effect from 10(-7) to 10(-5) M. Our results show that estradiol metabolites can influence the growth of vascular endothelial cells in the concentration range tested. While the antiproliferative action of 2-methoxyestradiol has been known for some time this study is the first to show the potential capacity of non-methylated metabolites of the A-ring metabolism in inhibiting endothelial proliferation. This may open up new clinical pharmacological aspects in the anti-angiogenetic treatment of tumors.

Urinary excretion of vasoactive markers following estrogen replacement therapy in postmenopausal women

OBJECTIVE

The effect of estradiol on the renal excretion of vasoactive substances was studied in postmenopausal women. The following markers surrogating an estrogenic effect on the cardiovascular system were measured: prostacyclin and thromboxane, cGMP, which reflects systemic nitric monoxide production, serotonin and relaxin.

METHODS

The effect of estradiol was compared using two clinical forms of administration, transdermal and oral, in two groups of 20 postmenopausal women each. The treatment was carried out for two and four weeks, respectively. Nocturnal urine was collected over 8 hours before and after estradiol treatment. The quantity of markers excreted during the experiment was determined.

RESULTS

Excretion of prostacyclin and thromboxane, calculated as prostacyclin/thromboxane ratio, was increased using both forms of administration. Both forms of treatment brought about only slight non-significant changes in renal cGMP excretion compared with values before treatment. The production of serotonin and relaxin was only increased using transdermal treatment.

CONCLUSION

The resulting data show that estradiol replacement in postmenopausal women is able to increase renal excretion of various vasoactive substances implying a vasodilative effect of estrogen. This was seen, both in transdermal and oral administration, transdermal application having a more pronounced effect on the markers than oral administration.

Consumers are driving change

Should hospitals, which have traditionally defined payers and physicians as their customers, be concerned about issues of consumer interest? Evidence is growing that consumers are redefining the formula of success for hospitals. Consider these consumer-oriented changes made by some of the nation's leading health organizations.

[Leiomyosarcoma of the spermatic cord with paraneoplastic beta-hCG production]

Leiomyosarcomas of the spermatic cord are very rare tumours, which, in contrast to the rhabdomyosarcoms of childhood, occur almost exclusively in adults. We report the case of a 51-year-old male patient with a metastasizing, pleomorphic leiomyosarcoma of the spermatic cord. The tumour showed paraneoplastic secretion of beta-chain human chorionic gonadotropin (beta-hCG), as was demonstrated by the elevated levels of beta-hCG in serum samples and by immunohistochemistry. Histologically, the tumour was a high-grade leiomyosarcoma which showed an aggressive course with pulmonary metastases appearing 4 months after primary surgery. A concomitant rise in the serum levels of beta-hCG was also noted at this time. A wide spectrum of tumours with a choriocarcinomatous component or paraneoplastic production of beta-hCG has been described, the vast majority being carcinomas. Only two leiomyosarcomas producing beta-hCG have so far been reported. The paraneoplastic production of beta-hCG should prevent confusion with germ-cell tumours, especially teratomas.

The effect of food on the bioavailability of dolasetron mesylate tablets

Anzemet (dolasetron mesylate) is being developed for the prevention of chemotherapy-induced emesis and postoperative nausea and vomiting. Twenty-four healthy male subjects were orally dosed with dolasetron mesylate, 200 mg, after either an overnight fast or a high-fat breakfast. The ratio of the mean area under the plasma concentration-time curve of the reduced active metabolite (MDL 74,156) to infinity (AUC(0-infinity)) values in fed compared to fasting subjects was 86.3% with a 90% confidence interval for the ratio within (80, 125)%, indicating bioequivalence. The ratio of the mean MDL 74,156 maximum plasma concentration (Cmax) values was 70.6% in fed versus fasted subjects. The time to Cmax was statistically significantly longer after the high-fat breakfast (mean values, 1.11 h fasting and 1.80 h fed), probably due to delayed gastric emptying. It may be concluded that, although the rate of absorption was somewhat delayed, the extent of absorption did not change significantly when dolasetron mesylate was given with food.

Use of hospital-based services in the first three months of life: impact of an early discharge program

OBJECTIVE

To assess the effect of an early discharge program on the use of hospital-based health care services in the first 3 months of life.

DESIGN

Retrospective cohort study.

SETTING

Metropolitan university hospital and a children's hospital.

PATIENTS

Term infants cared for in a single term nursery, before and after implementation of an early discharge program.

INTERVENTION

Early discharge program.

METHODS

Linking of the birth hospital and the children's hospital records and chart review.

OUTCOME MEASURES

Pattern of emergency department visits and rehospitalizations in the first 3 months of life.

RESULTS

The early discharge group had a shorter stay, 32 +/- 21 hours (mean +/- SD) than the control group (48 +/- 22 hours). There was no effect of early discharge on mean age at rehospitalization, rehospitalization rate, or reason for rehospitalization. Twenty-eight percent of infants in both study and control groups had at least one emergency department visit by 3 months of age. There was no difference between study and control groups in mean age or frequency of emergency department visits. Maternal age and race had a significant effect on the odds of visiting the emergency department. For any maternal age, nonwhite mothers were more likely to visit the emergency department.

CONCLUSIONS

Early discharge of newborn infants to inner city parents can be accomplished without increasing hospital-based resource use in the first 3 months of life provided coordinated postdischarge care and home visiting services are available.

CCK antagonist pre-treatment inhibits meal-enhanced drug absorption in dogs

Duodenal administration of casein and oleate increased plasma levels from oral administration of a poorly water-soluble antiepileptic drug as compared to duodenal glucose and saline in a canine model. Pre-treatment with intravenous MK-329, benzodiazepine CCK A-receptor antagonist, blocked the duodenal oleate effect on drug plasma levels in a single dog preliminary study. In a follow-up study, oral drug co-administration with Intralipid increased drug plasma levels as compared to drug co-administration with a noncaloric equivalent-volume load in seven dogs. Pre-treatment with MK-329 reduced drug plasma levels from co-administration with Intralipid toward fasted-state values. While increased drug solubility in the lipid vehicle might have been projected to account for the fed-state effect in the oral studies, the gut peptide inhibitor studies suggest that biliary secretion plays a major role in promoting the dissolution and subsequent absorption of this lipophilic drug. The data also support the hypothesis that meal-enhanced pancreatic secretion provides a greater fluid volume for drug dissolution in the small intestine. An increase in the extent of drug dissolution in the stomach, as a result of meal prolongation of gastric residence time, does not appear to contribute substantially to fed-state increases in drug plasma levels from oral drug co-administration with a lipid meal.

Continuity and change in preferred provider organizations

This paper presents the results from a national survey of preferred provider organizations (PPOs) that was conducted in 1988. It is based on telephone interviews conducted by the authors with executives in over 170 PPOs in the United States. We compare the survey results with those obtained from similar surveys conducted in 1985 and 1986, allowing us to assess the extent to which PPOs have grown and changed. We found that PPOs have continued to grow at an extremely rapid rate. During the Summer and Fall of 1988, the time in which the survey took place, 37.6 million people were eligible to use PPO benefits, compared to the 16.5 million figure we obtained two years earlier. We did not find, however, that PPOs are moving in the direction of providing more innovative forms of health care cost containment. Most PPOs still rely on discounts from providers and utilization review to achieve savings. There is little trend towards using incentive reimbursement techniques and choosing preferred providers that have shown themselves to be cost-efficient. We conclude that in the coming years PPOs must demonstrate the ability to control rising health care costs. To accomplish this, they will need to put more pressure on providers to use resources more sparingly. Otherwise, they may lose their market share to other forms of managed care.