Author + information
- Received July 25, 2002
- Revision received March 19, 2003
- Accepted March 27, 2003
- Published online July 2, 2003.
- Sergio Castellani, MD*,* (, )
- Rita Paniccia, PhD*,
- Claudia Di Serio, PhD†,
- Giuseppe La Cava, MD‡,
- Loredana Poggesi, MD*,
- Stefano Fumagalli, MD†,
- Gian Franco Gensini, MD* and
- Gian Gastone Neri Serneri, MD*
- ↵*Reprint requests and correspondence:
Prof. Sergio Castellani, Dipartimento di Area Critica Medico-Chirurgica, Sezione Clinica Medica Generale e Cardiologia, Viale Morgagni 85, 50134, Firenze, Italy.
Objectives The aim of this study was to evaluate whether thromboxane inhibition can favorably affect renal perfusion and clinical conditions in patients affected by severe heart failure.
Background The renal formation of the vasoconstrictor thromboxane A2(TxA2) is increased during cardiac failure.
Methods By oral administration of picotamide (a renal TxA2synthase and TxA2/prostaglandin H2receptor inhibitor), we blocked renal TxA2. Fourteen patients in New York Heart Association functional class IV were studied according to a randomized, double-blinded, cross-over design. Each of the two eight-day periods of testing was preceded by a three-day period during which certain vasoactive medications were stopped.
Results Daily 24-h total urinary thromboxane B2(TxB2), the stable metabolite of TxA2, dropped at the end of picotamide treatment (p < 0.01 vs. baseline). Compared with placebo, effective renal plasma flow and the glomerular filtration rate increased (p < 0.01 and p < 0.05, respectively), thus leading to a significant decrease in the filtration fraction (p < 0.01). Renal vascular resistance decreased consistently (p < 0.01). In all patients, picotamide treatment was associated with an increase in diuresis and natriuresis (p < 0.001 vs. baseline). Plasma creatinine decreased (p < 0.05 vs. baseline). Patients also showed improvement in several clinical parameters, including a significant decrease in both pulmonary and venous pressure (p < 0.01 vs. baseline).
Conclusions These results indicate that renal thromboxane formation plays an important role in renal vascular resistance in patients with severe heart failure, such as those described in the present study. Inhibition of TxA2improves renal hemodynamics and kidney function and favorably affects indexes of cardiac performance.
Chronic congestive heart failure (CHF) is characterized by the activation of neurohormonal vasoconstrictive pathways. These include the sympathetic nervous system, the renin-angiotensin system, vasopressin, and endothelin (1–4). The neurohormonal effects contribute to maintaining effective plasma volume by stimulating the tubulo-glomerular feedback mechanism (2). This mechanism is associated with a progressive decline in renal plasma flow and an increase in the filtration fraction (FF) due to selective constriction of the efferent arterioles (5–7)to preserve glomerular filtration. The increased formation of vasoconstrictor mediators is associated with enhanced formation of systemic and renal vasodilative factors, such as cardiac natriuretic peptides and renal prostaglandin (PG)E2and PGI2(1,8). Renal PGE2and PGI2counteract vasoconstriction, hence contributing to the maintenance of intrarenal circulatory homeostasis (9). A progressive increase in renal production of vasoconstrictor eicosanoids, such as PGF2αand thromboxane A2(TxA2), occurs in the more advanced stages of CHF (10). This increase most likely results in a decrease in renal blood flow, as suggested by experimental studies (11–13). Thromboxane A2promotes tubulo-glomerular feedback in microperfused single-nephron preparations (14). It causes renal vasoconstriction when injected in animals (11), thus resulting in increased sodium and water retention.
Under physiologic conditions, TxA2renal production in humans is low and does not play any role in renal hemodynamics (15). In pathophysiologic conditions such as CHF, renal function can remarkably worsen in the presence of large and persistent increases in renal TxA2formation (10). To investigate renal TxA2formation and its role in renal hemodynamics in patients with severe CHF, we used a TxA2synthase and TxA2/PGH2receptor inhibitor called picotamide (16). Several clinical trials have shown that picotamide can effectively block TxA2activity, thus resulting in good clinical effects (17–19).
We investigated 14 patients in New York Heart Association (NYHA) functional class IV (10 males and 4 females; mean age 70 ± 10 years). The demographic and clinical characteristics of the patients are reported in Table 1. Patients enrolled in the study had chronic CHF due to idiopathic dilated cardiomyopathy (n = 3), ischemic dilated cardiomyopathy (n = 10), and mitral regurgitation (n = 1). The patients with ischemic dilated cardiomyopathy had a previous (more than six months before the start of the study) myocardial infarction; none of them reported any episodes of angina during testing or for six months before the start of the study. No patients had clinically detectable edema or pulmonary rales. Patients who were enrolled in this study required hospitalization due to the inability for home treatment of NYHA functional class IV CHF. Because aspirin and nonsteroidal anti-inflammatory drugs can interfere with the effects of picotamide, patient selection criteria were also based on the exclusion of the use of aspirin or angiotensin-converting enzyme (ACE) inhibitors.
Patients with diabetes, liver or renal disease, malignancies, or serum creatinine levels ≥2.0 mg/dl were excluded from the study. The investigation followed the principles of the Declaration of Helsinki. All patients gave their informed, written consent to participate in the study after a clear and thorough explanation of the program had been presented.
The study was designed as a randomized, double-blinded, placebo-controlled, cross-over study. The patients were treated with digitalis, diuretics, and nitrates. Before hospitalization, two patients with atrial fibrillation had been on oral anticoagulants; during the study, they continued this treatment (international normalized ratio 2.5 to 3.0). Administration of digitalis, vasodilators, and diuretics was continued at the pre-established optimal dosage. Given that ACE inhibition can stimulate prostaglandin and thromboxane formation via bradykinin activation (20,21), ACE inhibitor administration was discontinued during the 48 h before the beginning the study and during the study. To prevent any possible deterioration in the clinical picture due to the withholding of ACE inhibitors, the enrolled patients were kept under strict, close, continuous clinical surveillance during the whole trial period. Only those patients whose conditions had remained stable after the first 12 h after ACE inhibitor discontinuation were admitted into the study. Patients remained under surveillance during the entire period of hospitalization until the end of the study. Initially, 20 patients had been selected, but only 14 were admitted into the study (6 patients had shown signs of instability during the 12-h pretreatment evaluation period). To obtain an index of the modifications of the patients’ clinical status (during testing), we evaluated the variations in tachycardia, breathlessness, and the occurrence of pulmonary rales and clinically detectable edema (before, during, and after the testing period). Evaluation was carried out twice by two cardiologists, neither of whom had any knowledge of the laboratory results, tests carried out, and/or treatment. During the total hospital period, the subjects were kept on a daily diet that contained 100 mmol of sodium chloride and 60 to 80 mmol of potassium. Dietary compliance with sodium intake was checked by measuring urinary sodium. Daily fluid intake was checked so that a constant water balance was maintained during the three-day pretreatment period. Patients received two tablets per day of either placebo or picotamide (600 mg, orally). After eight days of treatment, the patients underwent a three-day washout period before crossing over to the other treatment arm for a second eight-day period. The urine sample from the first day of the two eight-day periods was used to determine baseline values. On the second, fourth, sixth, and eighth days of both treatment periods, serum and urinary electrolytes, 24-h urinary volume, and 24-h urinary thromboxane excretion were measured. On the eighth day of both treatment periods, we studied renal hemodynamics. This experimental design allowed for a comparison between the treatment and placebo effects on renal hemodynamic parameters, without the need to expose the patients to any additional radioisotopic clearance studies.
Systemic hemodynamic parameters
Blood pressure and heart rate were recorded on an average of six times a day (at 6:00, 8:00, and 12:00 amand 4:00, 8:00, and 12:00 pm). Results were used as an index of the 24-h study period. They were also used to control levels of hypotension (a symptomatic sudden and/or persistent decline in arterial systolic blood pressure <105 mm Hg) and/or orthostatic hypotension (i.e., a drop in systolic blood pressure ≥20 mm Hg or in diastolic blood pressure ≥10 mm Hg) (22). Mean pulmonary artery pressure (mPAP) was measured by Doppler echocardiography on day 0 before the start of each eight-day period and on days 2, 4, 6, and 8 of each testing period. In one patient not showing tricuspid insufficiency, the pulmonary acceleration time was used to estimate mPAP, given that the two have been shown to be directly and strongly correlated (23). If tricuspid regurgitation was present, the atrioventricular pressure gradient was used to estimate mPAP, as described elsewhere (24). Venous pressure was obtained directly (25)through an antecubital vein.
Renal excretory function
Blood urea nitrogen, serum creatinine, serum and urinary sodium, and water excretion were monitored every other day throughout the hospital period. Urinary and serum sodium and potassium were measured by using a flame photometer (Instrumentation Laboratory, Lexington, Massachusetts).
At the end of the two periods of the study, two radioisotopes were rapidly injected into an antecubital vein with a 20-gauge needle, (iodine-125 [iothalamate] and iodine-131 [hippurate], both at a dose of 1 μCi/kg in a total volume of 9 ml). Six-milliliter blood samples were withdrawn from the other antecubital vein at 2, 5, 10, 15, 30, 45, 60, 90, 120, 150, and 180 min after isotopic injection. Blood samples were immediately centrifuged and a 3-ml plasma volume was measured for 5 min with a gamma ray, well-counter spectrometer (Packard Instrument Company, Palo Alto, California). Effective renal plasma flow (ERPF) and the glomerular filtration rate (GFR) were measured according to the methods described by Sapirstein et al. (26). The obtained values were normalized to a standard body surface area of 1.73 m2.
Immediately after collection, the urine was frozen and stored at −20°C until extraction and purification procedures could be carried out. Renal thromboxane B2(TxB2) was measured by radioimmunoassay (RIA), as reported elsewhere (27). A tritium tracer RIA was performed after allowing four to five months for the radioactivity of the iodine-125 and -131 in the urine to decay to background levels so as not to interfere with the TxB2RIA (28). Furthermore, before processing the urine for RIA, each sample was placed in a gamma counter to rule out any residual radioactivity. The data are expressed as ng/g of 24-h urinary creatinine to normalize the detected values according to the changes in renal function.
Continuous variables are expressed as the mean value ± SD. Discrete variables are expressed as percentages. Comparisons between groups in basal conditions were performed by using the Student ttest. Before analyzing the effects of the treatments on the different clinical parameters examined, we compared the baseline values with those obtained after the washout period, at the beginning of the second part of the study. We first tested for the presence of any differences within the same group, then between groups with the paired ttest and Student ttest, respectively. Given the lack of any dissimilarity (data not shown), we performed a general linear model for repeated measures to assess the effects of picotamide and placebo; patient data were pooled according to treatment. Comparisons between each observation time within each treatment group were performed in the same model specifying that the baseline value was reference one and testing for the presence of any statistically significant differences at days 2, 4, 6, and 8 (“simple” method) (29). Finally, we searched for differences between groups at each step of the study by using the Student ttest. A value of p < 0.05 was considered statistically significant. Given that 20 patients were initially selected, but only 14 were admitted to the study after ACE inhibitor suspension, we performed a retrospective power analysis to evaluate whether the sample size was appropriate. Assuming a mean value of TxB2excretion of 900 ± 122 ng/24 h for the 14 subjects enrolled, we built a model with a two-tailed alpha level of 0.05, obtaining a precision in estimating a difference in the population of 67 ng/24 h. According to this estimate, for the difference of 468 ng/24 h in TxB2excretion observed in the study, the power was >99.9%.
Effects of placebo administration
No adverse clinical events or severe hypotension occurred during placebo administration. During placebo administration, no change was found in urinary TxB2excretion (926 ± 154 vs. 878 ± 126 ng/g urinary creatinine [UCr]/24 h [first vs. eighth day], p = NS) (Fig. 1). Because renal and hemodynamic parameters at baseline were similar both before treatment and in the placebo periods, the three-day washout period appeared sufficient to abolish the effects of picotamide. There were not significant changes during the eight days on placebo.
Effects of picotamide treatment
A significant decrease in urinary TxB2was observed during the period of active treatment (p < 0.001) (Fig. 1). By the sixth day, urinary TxB2significantly dropped with respect to baseline (557 ± 138 vs. 900 ± 122 ng/g UCr/24 h; −38.2%; p < 0.01) and reached its minimum on the eighth day of picotamide treatment (432 ± 117 vs. 900 ± 122 ng/g UCr/24 h; −52.1%; p < 0.01). Moreover, compared with placebo, a significant decrease in urinary TxB2was observed on the sixth day (557 ± 138 vs. 893 ± 148 ng/g UCr/24 h; −37.7%; p < 0.01) and on the eighth day of picotamide treatment (432 ± 117 vs. 878 ± 126 ng/g UCr/24 h; −50.8%; p < 0.01) (Fig. 1).
The modifications in thromboxane excretion were accompanied by significant changes in renal hemodynamics, as shown in Figure 2. When comparing active treatment with placebo treatment, picotamide administration increased ERPF (+31.4%, p < 0.002) more than GFR (+7.9%, p < 0.05 vs. placebo), thus leading to a reduction in FF (−16%, p < 0.01 vs. placebo). The overall renal vascular resistance (RVR) decreased substantially (−26.8%, p < 0.01 vs. placebo) (Fig. 2).
Table 2reports the effect of active treatment on the different parameters assayed. We noted in all patients a decrease in both mPAP (−20.6% vs. baseline, p < 0.001) and venous pressure (−37.5%, p < 0.01); this effect was associated with clinical improvement, as indicated by a reduction in symptoms and increased sodium and water excretion. Blood pressure did not vary during treatment, whereas the heart rate significantly decreased. Picotamide treatment resulted in clinical improvement leading to dyspnea reduction and a decrease in tachycardia. Body weight significantly decreased on day 8 (−3.5%, p < 0.01 vs. day 1) of picotamide treatment. No bleeding, hypotension, or other adverse clinical events occurred during picotamide administration.
Thromboxane inhibition by picotamide and renal function
In severe CHF, picotamide administration inhibited renal thromboxane production in our patients who were not exposed to beta-adrenergic blockade or angiotensin activity suppression. The inhibition of renal thromboxane was consistent with the drop in urinary TxB2, which is a metabolite of renal TxA2formation and not of platelet TxA2(30). Blockade of the renal effects of TxA2led to a partial recovery of renal hemodynamics, as indicated by the 31.4% increase in ERPF and 26.8% decline in RVR. The renal resistance represents a very important part of total peripheral vascular resistance, and its increase occurs soon during the onset of CHF, thus leading to augmentation of the cardiac work load. The increases in ERPF and GFR following renal TxA2inhibition most probably indicate that the exaggerated TxA2release in CHF participates in constriction of the renal resistance vessels and likely with selective activity on afferent arterioles. This interpretation is in agreement with a recent study providing evidence that the afferent glomerular arteriole is the vascular segment that is preferentially constricted by TxA2(12). Inhibition of TxA2can lead to renal vasodilation, even in the absence of ACE inhibitors, highlighting the importance of the overproduction of renal thromboxane in the pathophysiology of CHF. The effects of picotamide were investigated in a cross-over experimental design where all patients were under controlled daily sodium intake, and the therapy remained unmodified during the whole study. The only variable was picotamide versus placebo administration. The lack of variations in renal function during placebo and the hemodynamic changes observed after picotamide administration should most probably not be interpreted as regression to the mean but most likely as real modification and improvement.
The improvement in renal function appeared to be related to the drop in pulmonary artery pressure and venous pressure, both associated with inhibition of renal thromboxane activity. The functional effects of TxA2inhibition became evident on the eighth day of picotamide administration. This latency seems to be dependent on the cumulative activity peculiar to this drug (18).
TxA2and renal pathophysiology in heart failure
In CHF, sympathetic activation and enhanced activity of the renin-angiotensin system, as well as vasopressin, play a major role in maintaining perfusion pressure and blood volume (2). An increase in renal formation of vasodilating PGE2and PGI2occurs in the early stages of heart failure (8)and can modulate the effects of vasoconstrictor mediators (1,9). This increase in vasodilating eicosanoid formation helps to sustain adequate renal glomerular filtration (1). With the progressive reduction in renal blood flow in the advanced phases of CHF (i.e., patients in NYHA functional class III and especially class IV), TxA2formation is enhanced (10), thus increasing the RVR. The increased TxA2formation results in the upset of the fine balance between renal vasoconstrictor and vasodilating systems. The inhibition of TxA2activity causes an increase in ERPF with decrease in RVR, resulting in enhanced water and sodium excretion. Furthermore, the inhibition of TxA2synthesis augments the availability of arachidonic acid, which is used for the synthesis of vasodilating PGs (31,32). In addition to its influence on renal hemodynamics, TxA2can potentiate tubulo-glomerular feedback by activating the sensory component in the macula densa (32).
In summary, the favorable effects of TxA2inhibition on renal function in patients with advanced CHF may be the consequence of three coexisting and interrelated mechanisms: 1) inhibition of the direct vasoconstrictive effect of renal TxA2on the afferent arterioles and mesangial cells; 2) enhancement of vasodilating PG synthesis due to the increased availability of arachidonic acid endoperoxides (PGG2and PGH2) for renal formation of PGE2and PGI2; and 3) reduction of the tubulo-glomerular feedback activation.
To our knowledge, no study has addressed the problem of picotamide action directly on the single nephron. Therefore, we cannot exclude the possibility that the natriuretic, kaliuretic, and diuretic effects of picotamide may have been partly due to some direct picotamide tubular action.
It is not possible to transfer the results obtained in our trials to all patients with heart failure, especially those who are usually treated with aspirin and/or ACE inhibitors. Such patients were excluded from the study, as were those patients with serious renal dysfunction (i.e., patients with creatinine >2 mg/dl). This latter complication can confound the evaluation of renal function after picotamide administration.
Our results suggest that increased renal TxA2formation is an important component of the complex pathophysiology of heart failure and that inhibition of renal thromboxane activity notably improves kidney function. However, no conclusions can be drawn regarding the use of TxA2inhibitors in the treatment of CHF. To do so, clinical trials must be expressly designed to address this aspect.
We wish to thank Prof. John C. McGiff, Department of Pharmacology, New York Medical College, Valhalla, New York, and Prof. Christopher S. Wilcox, Division of Nephrology and Hypertension, Georgetown University Medical Center, Washington, DC, for their most appreciated suggestions and assistance in preparing the manuscript. We also thank Miss Brunella Bandinelli for her skillful technical assistance.
☆ This research was part of the National Research Project, “Insufficienza Cardiaca.” Funding was provided by the Ministero dell’Università e della Ricerca Scientifica e Tecnologica (MURST) once the project had been approved by MURST (12.02.939 and 12.027339).
- angiotensin-converting enzyme
- congestive heart failure
- effective renal plasma flow
- filtration fraction
- glomerular filtration rate
- mean pulmonary artery pressure
- New York Heart Association
- renal vascular resistance
- thromboxane A2
- thromboxane B2
- urinary creatinine
- Received July 25, 2002.
- Revision received March 19, 2003.
- Accepted March 27, 2003.
- American College of Cardiology Foundation
- Stewart D.J.,
- Cernacek P.,
- Costello K.B.,
- Rouleau J.L.
- Dzau V.J.,
- Colucci W.S.,
- Hollenberg N.K.,
- Williams G.H.
- Ichikawa I.,
- Pfeffer J.M.,
- Pfeffer M.A.,
- Hostetter T.H.,
- Brenner B.M.
- ↵Dzau VJ, Packer M, Lilly LS, Swartz SL, Hollenberg NK, Williams GH. Prostaglandins in severe congestive heart failure: relation to activation of the renin-angiotensin system and hyponatremia. N Engl J Med 984;310:347–52
- Cirino M.,
- Morton H.,
- MacDonald C.,
- Hadden J.,
- Ford-Hutchinson A.W.
- Schnackenberg C.G.,
- Welch W.J.,
- Wilcox C.S.
- Zipser R.D.
- the ADEP Group,
- Balsano F.,
- Violi F.
- Giustina A.,
- Perini P.,
- Desenzani P.,
- et al.
- Cocozza M.,
- Picano T.,
- Oliviero U.,
- Russo N.,
- Coto V.,
- Milani M.
- Masaki K.H.,
- Schatz I.J.,
- Burchfiel C.M.,
- et al.
- Kitabatake A.,
- Inoue M.,
- Asao M.,
- et al.
- Currie P.J.,
- Seward J.B.,
- Chang K.L.,
- et al.
- Sapirstein L.A.,
- Vidt D.G.,
- Mandel M.J.,
- et al.
- ↵GLM Repeated Measures. In: SPSS Advanced Models 9.0. Chicago, IL: SPSS Inc., 1999:15–32
- Fitzgerald G.A.,
- Pedersen A.K.,
- Patrono C.
- Zoja C.,
- Perico N.,
- Corna D.,
- et al.