Splanchnic vasoregulation during mesenteric ischemia and reperfusion in pigs

Organ perfusion during voluntary pulmonary hyperinflation; a magnetic resonance imaging study

Pulmonary hyperinflation is used by competitive breath-hold divers and is accomplished by glossopharyngeal insufflation (GPI), which is known to compress the heart and pulmonary vessels, increasing sympathetic activity and lowering cardiac output (CO) without known consequence for organ perfusion. Myocardial, pulmonary, skeletal muscle, kidney, and liver perfusion were evaluated by magnetic resonance imaging in 10 elite breath-hold divers at rest and during moderate GPI. Cardiac chamber volumes, stroke volume, and thus CO were determined from cardiac short-axis cine images. Organ volumes were assessed from gradient echo sequences, and organ perfusion was evaluated from first-pass images after gadolinium injection. During GPI, lung volume increased by 5.2 ± 1.5 liters (mean ± SD; P < 0.001), while spleen and liver volume decreased by 46 ± 39 and 210 ± 160 ml, respectively (P < 0.05), and inferior caval vein diameter by 4 ± 3 mm (P < 0.05). Heart rate tended to increase (67 ± 10 to 86 ± 20 beats/min; P = 0.052) as right and left ventricular volumes were reduced (P < 0.05). Stroke volume (107 ± 21 to 53 ± 15 ml) and CO (7.2 ± 1.6 to 4.2 ± 0.8 l/min) decreased as assessed after 1 min of GPI (P < 0.01). Left ventricular myocardial perfusion maximum upslope and its perfusion index decreased by 1.52 ± 0.15 s(-1) (P < 0.001) and 0.02 ± 0.01 s(-1) (P < 0.05), respectively, without transmural differences. Pulmonary tissue, spleen, kidney, and pectoral-muscle perfusion also decreased (P < 0.05), and yet liver perfusion was maintained. Thus, during pulmonary hyperinflation by GPI, CO and organ perfusion, including the myocardium, as well as perfusion of skeletal muscles, are reduced, and yet perfusion of the liver is maintained. Liver perfusion seems to be prioritized when CO decreases during GPI.

Hypoxic hepatitis - epidemiology, pathophysiology and clinical management

Hypoxic hepatitis (HH), also known as ischemic hepatitis or shock liver, is characterized by centrilobular liver cell necrosis and sharply increasing serum aminotransferase levels in a clinical setting of cardiac, circulatory or respiratory failure. Nowadays it is recognized as the most frequent cause of acute liver injury with a reported prevalence of up to 10% in the intensive care unit. Patients with HH and vasopressor therapy have a significantly increased mortality risk in the medical intensive care unit population. The main underlying conditions contributing to HH are low cardiac output and septic shock, although a multifactorial etiology is found in the majority of patients. HH causes several complications such as spontaneous hypoglycemia, respiratory insufficiency due to the hepatopulmonary syndrome, and hyperammonemia. HH reverses after successful treatment of the basic HH-causing disease. No specific therapies improving the hepatic function in patients with HH are currently established. Early recognition of HH and its underlying diseases and subsequent initiation of therapy is of central prognostic importance. The purpose of this review is to provide an update on the epidemiology, pathophysiology, and diagnostic and therapeutic options of HH.

Is impaired hepatic arterial buffer response a risk factor for biliary anastomotic stricture in liver transplant recipients?

BACKGROUND

Blood flow to the liver is partly maintained by the hepatic arterial buffer response (HABR), which is an intrinsic autoregulatory mechanism. Temporary clamping of the portal vein (PV) results in augmentation in hepatic artery flow (augHAF). Portal hyperperfusion impairs HAF due to the HABR in liver transplantation (LT). The aim of this study is to examine the effect of the HABR on biliary anastomotic stricture (BAS).

METHODS

In 234 cadaveric whole LTs, PV flow (PVF), basal HAF, and augHAF were measured intra-operatively after allograft implantation. All recipients with a vascular complication were excluded. Buffer capacity (BC) was calculated as (augHAF - basal HAF)/PVF to quantify the HABR. Recipients were divided into 2 groups based on their BC: low BC (<0.074; n = 117) or high BC (> or =0.074; n = 117).

RESULTS

Of the 234 recipients, 23 (9.8%) had early BAS (< or =60 days after LT) and 18 (7.7%) had late BAS (>60 days after LT). The incidence of late BAS and bile leakage was similar between the groups; however, the incidence of early BAS in the low BC group was greater than that in the high BC group (15% vs 5.1%; P = .0168). In the multivariate analysis, low BC (P = .0325) and bile leakage (P = .0002) were found to be independent risk factors affecting early BAS.

CONCLUSION

Recipients with low BC who may have impaired HABR are at greater risk of early BAS after LT. Intraoperative measurements of blood flow help predict the risk of BAS.

Fetal superior mesenteric artery: longitudinal reference ranges and evidence of regulatory link to portal liver circulation

OBJECTIVE

To establish longitudinal reference ranges for the fetal superior mesenteric artery (SMA) flow velocity and pulsatility index (PI(SMA)). Also to examine the hemodynamic relationship to venous liver perfusion and umbilical flow distribution in the liver, to other splanchnic arteries, and more generally to the middle cerebral and umbilical artery.

METHODS

Prospective longitudinal study of 161 low-risk pregnancies using Doppler recordings including the SMA, repeated on 3-5 occasions at 3-5 weekly intervals. Umbilical venous flow was estimated, blood velocity in the shunt ductus venosus represented umbilico-caval (i.e. porto-caval) pressure gradient, and left portal vein blood velocity represented umbilical distribution within the liver. The correlation between PI(SMA) and the splenic and hepatic artery PI were analysed (PI(SA) and PI(HA)), and the association to middle cerebral and umbilical artery PI (PI(MCA) and PI(UA)) assessed.

RESULTS

Reference ranges for the SMA for gestational weeks 21-39 were based on 589 observations. Low impedance in the SMA (i.e. low PI(SMA)) was associated with low umbilical flow and porto-caval pressure gradient (i.e. <10th centile), and high distribution of umbilical flow to the right lobe (i.e. left portal vein blood velocity >90th centile). PI(SMA) correlated weakly with PI(SA) and PI(HA) (r=0.30, 95%CI 0.22-0.37, and r=0.39, 95%CI 0.27-0.51, respectively). PI(SMA) was positively associated with PI(MCA) and PI(UA).

CONCLUSION

We have provided longitudinal reference ranges for fetal SMA flow velocity and PI, and shown that the SMA, which perfuses the fetal gut, is also involved in the regulation of the liver perfusion.

Tonometry revisited: perfusion-related, metabolic, and respiratory components of gastric mucosal acidosis in acute cardiorespiratory failure

Mucosal pH (pHi) is influenced by local perfusion and metabolism (mucosal-arterial pCO2 gradient, DeltapCO2), systemic metabolic acidosis (arterial bicarbonate), and respiration (arterial pCO2). We determined these components of pHi and their relation to outcome during the first 24 h of intensive care. We studied 103 patients with acute respiratory or circulatory failure (age, 63+/-2 [mean+/-SEM]; Acute Physiology and Chronic Health Evaluation II score, 20+/-1; Sequential Organ Failure Assessment score, 8+/-0). pHi, and the effects of bicarbonate and arterial and mucosal pCO2 on pHi, were assessed at admission, 6, and 24 h. pHi was reduced (at admission, 7.27+/-0.01) due to low arterial bicarbonate and increased DeltapCO2. Low pHi (<7.32) at admission (n=58; mortality, 29% vs. 13% in those with pHi>or=7.32 at admission; P=0.061) was associated with an increased DeltapCO2 in 59% of patients (mortality, 47% vs. 4% for patients with low pHi and normal DeltapCO2; P=0.0003). An increased versus normal DeltapCO2, regardless of pHi, was associated with increased mortality at admission (51% vs. 5%; P<0.0001; n=39) and at 6 h (34% vs. 13%; P=0.016; n=45). A delayed normalization or persistently low pHi (n=47) or high DeltapCO2 (n=25) was associated with high mortality (low pHi [34%] vs. high DeltapCO2 [60%]; P=0.046). In nonsurvivors, hypocapnia increased pHi at baseline, 6, and 24 h (all P<or=0.001). In patients with initially normal pHi or DeltapCO2, outcome was not related to subsequent changes in pHi or DeltapCO2. Increased DeltapCO2 during early resuscitation suggests poor tissue perfusion and is associated with high mortality. Arterial bicarbonate contributes more to pHi than the DeltapCO2 but is not associated with mortality. Hyperventilation partly masks mucosal acidosis. Inadequate tissue perfusion may persist despite stable hemodynamics and contributes to poor outcome.

The immediate and sustained effects of volume challenge on regional blood flows in pigs

BACKGROUND

The postoperative assessment of volume status is not straightforward because of concomitant changes in intravascular volume and vascular tone. Hypovolemia and blood flow redistribution may compromise the perfusion of the intraabdominal organs. We investigated the effects of a volume challenge in different intra- and extraabdominal vascular beds.

METHODS

Twelve pigs were studied 6 h after major intraabdominal surgery under general anesthesia when clinically normovolemic. Volume challenges consisted of 200 mL rapidly infused 6% hydroxyethyl starch. Systemic (continuous thermodilution) and regional (ultrasound Doppler) flows in carotid, renal, celiac trunk, hepatic, and superior mesenteric arteries and the portal vein were continuously measured. The acute and sustained effects of the challenge were compared with baseline.

RESULTS

Volume challenge produced a sustained increase of 22% +/- 15% in cardiac output (P < 0.001). Blood flow increased by 10% +/- 9% in the renal artery, by 22% +/- 15% in the carotid artery, by 26% +/- 15% in the superior mesenteric artery, and by 31% +/- 20% in the portal vein (all P < 0.001). Blood flow increases in the celiac trunk (8% +/- 13%) and the hepatic artery (7% +/- 19%) were not significant. Increases in regional blood flow occurred early and were sustained. Mean arterial and central venous blood pressures increased early and decreased later (all P < 0.05).

CONCLUSIONS

A volume challenge in clinically euvolemic postoperative animals was associated with a sustained increase in blood flow to all vascular beds, although the increase in the celiac trunk and the hepatic artery was very modest and did not reach statistical significance. Whether improved postoperative organ perfusion is accompanied by a lower complication rate should be evaluated in further studies.

Effects of low abdominal blood flow and dobutamine on blood flow distribution and on the hepatic arterial buffer response in anaesthetized pigs

Low cardiac output impairs the hepatic arterial buffer response (HABR). Whether this is due to low abdominal blood flow per se is not known. Dobutamine is commonly used to increase cardiac output, and it may further modify hepatosplanchnic and renal vasoregulation. We assessed the effects of isolated abdominal aortic blood flow changes and dobutamine on hepatosplanchnic and renal blood flow. Twenty-five anesthetized pigs with an abdominal aorto-aortic shunt were randomized to 2 control groups [zero (n = 6) and minimal (n = 6) shunt flow], and 2 groups with 50% reduction of abdominal blood flow and either subsequent increased abdominal blood flow by shunt reduction (n = 6) or dobutamine infusion at 5 and 10 microg kg(-1) min(-1) with constant shunt flow (n = 7). Regional (ultrasound) and local (laser Doppler) intra-abdominal blood flows were measured. The HABR was assessed during acute portal vein occlusion. Sustained low abdominal blood flow, by means of shunt activation, decreased liver, gut, and kidney blood flow similarly and reduced local microcirculatory blood flow in the jejunum. Shunt flow reduction partially restored regional blood flows but not jejunal microcirculatory blood flow. Low-but not high-dose dobutamine increased gut and celiac trunk flow whereas hepatic artery and renal blood flows remained unchanged. Neither intervention altered local blood flows. The HABR was not abolished during sustained low abdominal blood flow despite substantially reduced hepatic arterial blood flow and was not modified by dobutamine. Low-but not high-dose dobutamine redistributes blood flow toward the gut and celiac trunk. The jejunal microcirculatory flow, once impaired, is difficult to restore.

Membrane microdialysis: Evaluation of a new method to assess splanchnic tissue metabolism

OBJECTIVE

Measuring peritoneal lactate concentrations could be useful for detecting splanchnic hypoperfusion. The aims of this study were to evaluate the properties of a new membrane-based microdialyzer in vitro and to assess the ability of the dialyzer to detect a clinically relevant decrease in splanchnic blood flow in vivo.

DESIGN

A membrane-based microdialyzer was first validated in vitro. The same device was tested afterward in a randomized, controlled animal experiment.

SETTING

University experimental research laboratory.

SUBJECTS

Twenty-four Landrace pigs of both genders.

INTERVENTIONS

In vitro: Membrane microdialyzers were kept in warmed sodium lactate baths with lactate concentrations between 2 and 8 mmol/L for 10-120 mins, and microdialysis lactate concentrations were measured repeatedly (210 measurements). In vivo: An extracorporeal shunt with blood reservoir and roller pump was inserted between the proximal and distal abdominal aorta, and a microdialyzer was inserted intraperitoneally. In 12 animals, total splanchnic blood flow (measured by transit time ultrasound) was reduced by a median 43% (range, 13% to 72%) by activating the shunt; 12 animals served as controls.

MEASUREMENTS AND MAIN RESULTS

In vitro: The fractional lactate recovery was 0.59 (0.32-0.83) after 60 mins and 0.82 (0.71-0.87) after 90 mins, with no further increase thereafter. At 60 and 90 mins, the fractional recovery was independent of the lactate concentration. In vivo: Abdominal blood flow reduction resulted in an increase in peritoneal microdialysis lactate concentration from 1.7 (0.3-3.8) mmol/L to 2.8 (1.3-6.2) mmol/L (p = .006). At the same time, mesenteric venous-arterial lactate gradient increased from 0.1 (-0.2-0.8) mmol/L to 0.3 (-0.3 -1.8) mmol/L (p = .032), and mesenteric venous-arterial Pco2 gradients increased from 12 (8-19) torr to 21 (11-54) torr (p = .005).

CONCLUSIONS

Peritoneal membrane microdialysis provides a method for the assessment of splanchnic ischemia, with potential for clinical application.

Hepatic Arterial Buffer Response Fails to Restore Hepatic Oxygenation After Temporary Liver Dearterialization in Canines

BACKGROUND

Hepatic artery thrombosis is a rare but extremely troublesome condition after liver transplantation. Recently, urgent arterial revascularization has been used as rescue therapy, leading to improved graft and patient survivals. Hepatic artery ligation produces a progressive reduction in portal vein blood flow. Theoretically, a hyperemic response may be expected following hepatic artery reperfusion (hepatic artery buffer response, HABR). In this study, we tested the hypothesis that HABR can maintain adequate liver oxygenation after temporary liver dearterialization.

METHODS

Seven dogs (19.7 +/- 1.2 kg) subjected to 60 minutes of hepatic artery occlusion were observed for 120 minutes thereafter. Systemic hemodynamics was evaluated through Swan-Ganz and arterial catheters, and splanchnic perfusion by portal vein and hepatic artery blood flows (PVBF and HABF) via an ultrasonic flowprobe. Liver enzymes (ALT and LDH) and systemic and hepatic oxygen delivery (DO2hepat) were calculated using standard formulae.

RESULTS

Hepatic artery occlusion induced a progressive reduction in PVBF and DO2hepat. A complete restoration of HABF after hepatic artery declamping was observed; however, the DO2hepat (33.3 +/- 5.9 to 16.5 +/- 5.9 mL/min) did not return to the baseline levels.

CONCLUSION

Temporary hepatic artery occlusion induced a progressive decrease in portal vein blood flow during ischemia, an effect that continued during the reperfusion period. The hepatic artery blood flow was promptly restored after declamping. However, HABR was not able to restore hepatic oxygen delivery to baseline levels during the reperfusion period.

Intraoperative direct measurement of hepatic arterial buffer response in patients with or without cirrhosis

The hepatic arterial buffer response (HABR) is an intrinsic regulatory mechanism of the hepatic artery (HA) that compensates for reductions in portal venous (PV) blood flow. Whether this response is maintained in patients with cirrhosis (LC) is unclear. The aim of the present study was to examine whether HABR is maintained in patients with LC using direct blood flow measurements. PV and HA blood flow were intraoperatively measured and compared in patients with (LC group, n = 39) or without (control group, n = 22) cirrhosis at baseline (baseline HABR) and after PV clamping (acute HABR) using an ultrasound transit-time flowmeter. In contrast to the proportional relationship between the baseline PV and HA blood flow observed in the control group, HA blood flow and the HA-PV flow ratio increased when PV blood flow decreased in the LC group, suggesting that the baseline HABR had already been activated. Acute HABR, evaluated by the absolute and relative changes in HA blood flow and by the buffer capacity, was blunted in the LC group (P < 0.001, P < 0.01, and P = 0.01, respectively). An association between the degree of acute HABR impairment and the level of baseline HABR activation (HA-PV flow ratio) could not be confirmed in the LC group. In conclusion, the baseline HABR appears to be continuously activated in patients with LC; this phenomenon probably results in the impairment of the acute HABR.

Oxygen extraction in pigs subjected to low-dose infusion of endotoxin after major abdominal surgery

BACKGROUND

Sepsis may impair O(2) extraction due to blood flow redistribution or decreased utilization of the available oxygen.

METHODS

We assessed the effect of endotoxemia on systemic and regional O(2) extraction and lactate handling in pigs, randomized to receive either endotoxin (0.4 microg kg(-1) h(-1); n = 10) or saline infusion (controls; n = 9) for 12 h.

RESULTS

High baseline regional and systemic O(2) extraction in the endotoxin group (median 56%, range 45-77%) and in the controls (67%, 49-72%) was maintained until the end of the experiment (endotoxin group: 60%, 50-71%; controls: 60%, 50-74%) despite hypotension and a decrease in stroke volume in endotoxic animals. Hepatic lactate exchange decreased during endotoxemia from 14 micromol kg(-1) min(-1) (range 10-28 micromol kg(-1) min(-1)) to 10 (range 3-15) micromol kg(-1) min(-1); P < 0.01), but remained stable in the controls, with 13 micromol min(-1) (4-18 micromol min(-1)) at baseline and 7 micromol min(-1) (3-17 micromol min(-1)) after 12 h of saline infusion.

CONCLUSIONS

The high and sustained oxygen consumption and oxygen extraction in this endotoxemic model speak against any major impairment of hepatosplanchnic or systemic oxygen extraction and oxidative metabolism. The reduced hepatic lactate exchange despite an unchanged hepatic lactate influx suggests altered metabolic activities independent of oxygen consumption.

Hepatosplanchnic blood flow control and oxygen extraction are modified by the underlying mechanism of impaired perfusion

OBJECTIVE

To assess the effects of low hepatosplanchnic blood flow on regional blood flow control and oxygenation.

DESIGN

Three randomized, controlled animal experiments.

SETTING

Two university experimental research laboratories.

SUBJECTS

Pigs of either gender.

INTERVENTIONS

Isolated abdominal blood flow reduction: An extracorporeal shunt with reservoir and roller pump was inserted between proximal and distal aorta in 11 pigs. Abdominal aortic blood flow was reduced by 50% by activating the shunt. Mesenteric ischemia: In seven pigs, superior mesenteric arterial flow was reduced to 4 mL.kg.min for 4 hrs. Cardiac tamponade: In 12 pigs, aortic blood flow was reduced by cardiac tamponade to 50 mL (moderate tamponade) and further to 30 mL.kg.min (severe tamponade) for 1 hr each. In each experimental condition, the same number of control animals was used.

MEASUREMENTS AND MAIN RESULTS

Abdominal blood flow reduction, acute mesenteric ischemia, and moderate tamponade resulted in a portal venous flow (QPV) reduction to 51 +/- 23%, 52 +/- 18%, and 61 +/- 25% (mean +/- sd) of baseline flow, respectively. During abdominal blood flow reduction, QPV and hepatic arterial flow (QHA) decreased proportionally, whereas in moderate tamponade and acute mesenteric ischemia QPV reduction was associated with an increase in QHA of 30 +/- 39% and 102 +/- 108%, respectively (p = .001 and .018). Prolonged mesenteric ischemia restored total hepatic blood flow (Qliver) completely. During all conditions, decreasing mesenteric oxygen consumption was partly prevented by increased mesenteric oxygen extraction (p < .001 for all conditions). In contrast, decreasing hepatic oxygen delivery was associated with increased oxygen extraction in tamponade (p = .009) but not in abdominal blood flow reduction.

CONCLUSIONS

Blood flow redistribution can restore Qliver totally when mesenteric blood flow is reduced selectively, partially when cardiac output is reduced, and not at all during abdominal blood flow reduction. Since hepatic oxygen extraction does not increase in abdominal blood flow reduction, hepatic oxygenation is at risk in this condition.

The effects of vasopressin on systemic and splanchnic hemodynamics and metabolism in endotoxin shock

We compared the effects of vasopressin and norepinephrine on systemic and splanchnic circulation and metabolism in endotoxin shock in pigs. Twenty-one pigs were randomized to endotoxin shock (Escherichia coli endotoxin infusion) (n = 6), endotoxin and vasopressin (VASO; n = 6), endotoxin and norepinephrine (NE; n = 6), and controls (n = 3). Endotoxin infusion was increased to induce hypotension, after which vasopressin or norepinephrine was started to keep systemic mean arterial blood pressure >70 mm Hg. Regional blood flows and arterial and regional lactate concentrations were measured. Tonometers with microdialysis capillaries were inserted into the stomach, jejunum, and colon. Systemic mean arterial blood pressure >70 mm Hg was achieved in the VASO and NE groups. Vasopressin decreased cardiac output, superior mesenteric artery, and portal vein blood flow, whereas hepatic arterial blood flow increased. Arterial lactate concentration increased from 2.0 mM (1.6-2.1 mM) to 4.7 mM (4.7-4.9 mM) (P = 0.007). Systemic and mesenteric oxygen delivery and consumption decreased and oxygen extraction increased in the VASO group. Vasopressin increased mucosal-arterial PCO(2) gradients in all three locations, whereas luminal lactate release occurred only in the jejunum. Animals in the NE group remained stable. Vasopressin reversed hypotension but decreased systemic and gut blood flow. This was associated with hyperlactatemia, signs of visceral dysoxia, and jejunal luminal lactate release.

IMPLICATIONS

Although vasopressin induces vasoconstriction in visceral region, its effects on splanchnic circulation and metabolism during septic-endotoxin shock are still poorly characterized. We evaluated the metabolic and hemodynamic effects of vasopressin and norepinephrine within the splanchnic area in porcine endotoxin shock.

Splanchnic blood flow in low-flow states

IMPLICATIONS

Insufficient splanchnic blood flow in critically ill patients is the result of a multitude of different diseases, treatment modalities and their interplay, and is associated with increased morbidity and mortality. A combination of diminished and heterogeneous mesenteric blood flow, impaired or exhausted regulatory mechanisms and adverse drug effects may coexist with normal systemic hemodynamics.

Mesenteric hemodynamic response to circulatory shock

PURPOSE OF REVIEW

The mesenteric hemodynamic response to circulatory shock is substantial and asymmetrical; the vasoconstrictive response disproportionately affects the mesenteric organs. The cardiac output is sustained partially, at no cost in nutrient flow to the mesenteric organs, by vasoconstriction of the mesenteric veins, resulting in the "autotransfusion" of up to 30% of the circulating blood volume into the systemic circulation.

RECENT FINDINGS

Hemorrhagic or cardiogenic shock also results in decreased perfusion pressure, prompting selective vasoconstriction of the mesenteric arterioles to maintain perfusion pressure of the vital organs, here at the selective expense of the mesenteric organs. Septic shock may be associated with increased or decreased mesenteric blood flow but is characterized by increased oxygen consumption, exceeding the capability of mesenteric oxygen delivery.

SUMMARY

The response to any of these conditions can, variably and unpredictably, cause hemorrhagic gastric stress erosions, nonocclusive mesenteric ischemia of the small bowel, ischemic colitis, ischemic hepatitis, acalculous cholecystitis, and/or ischemic pancreatitis. Injury to the mesenteric organs can also initiate the systemic inflammatory response syndrome and, consequently, multiple organ failure.

Clinical review: splanchnic ischaemia

Inadequate splanchnic perfusion is associated with increased morbidity and mortality, particularly if liver dysfunction coexists. Heart failure, increased intra-abdominal pressure, haemodialysis and the presence of obstructive sleep apnoea are among the multiple clinical conditions that are associated with impaired splanchnic perfusion in critically ill patients. Total liver blood flow is believed to be relatively protected when gut blood flow decreases, because hepatic arterial flow increases when portal venous flow decreases (the hepatic arterial buffer response [HABR]). However, there is evidence that the HABR is diminished or even abolished during endotoxaemia and when gut blood flow becomes very low. Unfortunately, no drugs are yet available that increase total hepato-splanchnic blood flow selectively and to a clinically relevant extent. The present review discusses old and new concepts of splanchnic vasoregulation from both experimental and clinical viewpoints. Recently published trials in this field are discussed.