First Report of Candidatus Phytoplasma solani Associated with Potato Plants in Greece.

In August 2013, potato plants (Solanum tuberosum) cv. Banba displaying symptoms resembling those caused by Candidatus Phytoplasma solani (potato stolbur phytoplasma) were observed in a 2-ha field in the area of the Peripheral Unit of Drama (northern Greece). The plants were 10 weeks old and their symptoms included reddening and upward rolling of leaflets, reduced size of leaves, shortened internodes, and aerial tuber formation. Incidence of affected plants was estimated to be 40% in the field. Four symptomatic potato plants were collected for laboratory testing of possible phytoplasma infection. From each of these four plants, total DNA was extracted from mid veins of reddish leaflets from apical shoot parts and of leaflets emerging from aerial tubers, using a phytoplasma enrichment procedure (1). A nested PCR using the phytoplasma universal 16S rRNA primer pairs: P1/P7 followed by R16F2n/R16R2 (3) amplified the expected ~1.2-kb 16S rDNA fragment in all four symptomatic potato plants. No amplification was observed with DNA similarly extracted from leaflets of asymptomatic potato plants of the same variety collected from an apparently healthy crop. One of the four 1.2-kb nested 16S rDNA PCR products was gel purified, cloned into the pGEM-T-easy plasmid vector (Promega, Madison, WI), and sequenced by Beckman Coulter Genomics (United Kingdom). At least twofold coverage per base position of the cloned PCR product was achieved. BLAST analysis showed that the obtained sequence of the PCR 16S rDNA product was: i) 100% identical to several GenBank sequences of Ca. P. solani strains, including strains detected previously in Greece infecting tomato (GenBank Accession No. JX311953) and Datura stramonium (HE598778 and HE598779), and ii) 99.7% similar to that of the Ca. P. solani reference strain STOL11 (AF248959). Furthermore, analysis by iPhyClassifier software showed that the virtual restriction fragment length polymorphism (RFLP) pattern of the sequenced PCR 16S rDNA product is identical (similarity coefficient 1.00) to the reference pattern of the 16SrXII-A subgroup (AF248959). The sequence of this PCR product was deposited in NCBI GenBank database under the accession no. KJ810575. The presence of the stolbur phytoplasma in all four symptomatic potato plants examined was further confirmed by nested PCR using the stolbur-specific STOL11 primers (3) targeting non-ribosomal DNA. Based on the observed symptoms in the field and laboratory molecular examinations, we concluded that the potato plants were infected by a Ca. P. solani related strain. The stolbur disease has been previously reported in Greece affecting tomato (2,5) and varieties of D. stramonium (4). To our knowledge, this is the first report of a Ca. P. solani related strain infecting a potato crop in Greece. As northern Greece is a center of potato production, the source of this pathogen is to be investigated. References: (1) U. Ahrens and E. Seemuller. Phytopathology 82:828, 1992. (2) A. S. Alivizatos. Pages 945-950 in: Proceedings of the 7th International Conference of Plant Pathogenic Bacteria. Academiai Kiado, Budapest, Hungary, 1989. (3) J. Jovic et al. Bull. Insectol. 64:S83, 2011. (4) L. Lotos et al. J. Plant Pathol. 95:447, 2013. (5) E. Vellios and F. Lioliopoulou. Bull. Insectol. 60:157, 2007.

Temporary stenting of acquired benign tracheoesophageal fistulas in critically ill ventilated patients.

BACKGROUND: To evaluate the use of esophageal stents for temporary sealing of acquired benign tracheoesophageal fistulas developed in critically ill, ventilated patients. METHODS: This is a retrospective analysis (1992-2003) of the data of 12 mechanically ventilated patients - six of them after major or multiple trauma - being intubated for a median of 30 days before they develop an acquired benign tracheoesophageal fistula. Five of them were in sepsis. Two types of stents were used: the Wilson-Cook esophageal balloon plastic stent in the first four cases and the Ultraflex covered self-expandable stent in the remaining eight. The total procedure was performed at bedside in the intensive care unit, with no special need for supplementary anesthesia or fluoroscopic control. RESULTS: Stent implantation was technically successful in all patients and fistula occlusion was achieved in every case. There was no stent migration and fistulas remained sealed until death or upon decision for removal. Nine patients died between 5 days and 2 months after stent placement, as a result of their diseases. Three patients were referred for fistula surgical repair 33, 36, and 43 days after stent placement. Before surgery the stents were easily removed under direct vision. CONCLUSION: Temporary closure of an acquired tracheoesophageal fistula developed in critically ill ventilated patients is an easy, bedside-applicable, safe, and effective palliative procedure, with no complications or mortality.

Percutaneous endoscopic gastrostomy after abdominal surgery.

BACKGROUND: Percutaneous endoscopic gastrostomy (PEG) has been established as a minimally invasive and safe procedure to provide nutritional support in patients unable to swallow food properly. However, a relative contraindication for the procedure is the existence of previous abdominal surgery. METHOD: We described our experience in performing PEG on 37 patients who had undergone surgery for upper abdominal diseases 10 days to 25 years previously. This group of 37 patients, 22 of whom had been subjected to laparotomy in the previous 2 weeks, was compared with 291 patients who had an intact abdomen regarding failure of the procedure to be performed, major and minor complications, and mortality. RESULTS: The two groups were found to be comparable: failure rates of 2.71% (1/37) and 1.72% (5/291), no major complications, no mortality, minor complication rates of 2.77% (1/36) and 2.44% (7/286). CONCLUSIONS: The PEG procedure is associated with minimal risk even in patients previously subjected to upper abdominal surgery, as long as transillumination of the stomach and finger palpation are seen clearly during endoscopy.

Sustained reduction of portal pressure by administration of octreotide-LAR in portal hypertensive rats.

Objective: Recent studies have demonstrated octreotide as being potentially beneficial for the long-term management of cirrhotic portal hypertension. However, its short-term effects, requiring subcutaneous administration at least twice daily, make it inconvenient, practically, for long-term therapy. Hence, the current study was undertaken to evaluate the effects of the new long-acting-release octreotide formulation (octreotide-LAR) on portal pressure in a prehepatic portal hypertensive rat model. Methods: 14 days after portal vein stenosis or sham operation, rats were treated with either octreotide-LAR (0.25 mg/kg body weight, subcutaneously) or vehicle, after the baseline portal pressure measurements were obtained in each group. On days 10, 15, 18, 21, 24, 27 and 30 post-treatment, eight rats from each sub-group (portal vein stenosis plus octreotide-LAR, or vehicle, sham-operated plus octreotide-LAR or vehicle) were subjected to portal vein pressure measurement. Results: portal vein-stenosed rats exhibited a significantly higher portal vein pressure than sham-operated rats (13.82+/-2.18 vs. 5.82+/-1.08 mmHg, P=0.0001). Octreotide-LAR treatment was found to significantly (P=0.001) reduce portal pressure in portal vein stenosed rats from the tenth post-injection day and thereafter to sustain reduction. Conclusion: the long-acting-release octreotide given in a single dose in a prehepatic portal hypertensive rat model diminishes and sustains diminished portal pressure for 20 days after the tenth post-injection day. These findings seem promising for the portal hypertensive patient and further clinical studies are necessary.

Endoscopic fibrin sealing of high-output non-healing gastrocutaneous fistulas after vertical gastroplasty in morbidly obese patients.

BACKGROUND: Fibrin glue was used in a various fields of surgery during the last 15 years, but its use has not been reported in bariatric surgery yet. METHODS: In 2 out of 215 morbidly obese patients who underwent vertical banded gastroplasty, a non-healing gastrocutaneus fistula (GCF) developed. In both patients sepsis occurred, caused by a leak of the posterior gastric wall, which was managed by means of an unsuccessful reoperation. After that, sepsis recurred, and a non-healing GCF developed. These GCF were managed endoscopically by the use of a fibrin sealant (Beriplast P 2 ml set, Behring) as a tissue adhesive. RESULTS: One injection was needed for the first case and six for the second in achieving full healing of the fistulas. No evidence of fistula was observed at gastroscopy 3 and 24 months after the end of therapy. CONCLUSIONS: Endoscopic use of human fibrin sealant is simple, safe, effective and in some cases life-saving. This is a therapeutic option in high output GCF in morbidly obese patients.

Deterioration of visceral perfusion caused by intra-abdominal hypertension in pigs ventilated with positive end-expiratory pressure.

Experimental studies and clinical experience suggest that the combination of positive end-expiratory pressure (PEEP) ventilation and intra-abdominal hypertension might alter splanchnic hemodynamics to a significantly greater degree than the effect of either of them alone. Therefore, we assessed the intestinal and hepatic hemodynamics in two steps of PEEP ventilation, adding tense pneumoperitoneum in a pig model. The hepatic artery, portal vein, and superior mesenteric artery blood flow, as well as the hepatic and intestinal mucosal microcirculation, and the hepatic pO2 and intestinal mucosal pH, were assessed before, then with 5 cmH2O and 10 cmH2O PEEP alone, and in combination with a 12-mmHg pneumoperitoneum, in ten domestic pigs. Statistical analysis of the hepatic and intestinal measurements revealed a significant decrease (P = 0.001) in all parameters in relation to the baseline, during the 5-cmH2O and 10-mmH2O PEEP ventilation period. The addition of 12 mmHg intra-abdominal pressure led to an extreme deterioration in all parameters (P = 0.001), in relation to both the baseline and the 10-cmH2O PEEP measurement. These findings demonstrate that PEEP and intra-abdominal hypertension act cumulatively on the abdominal viscera, producing conditions of extremely low hypoperfusion and ischemia.

The implication of nitric oxide in the process of bacterial translocation.

Since there is increasing evidence indicating nitric oxide [NO] would play a role in sepsis, we decided to investigate whether this multifaceted mediator is directly implicated in the process of bacterial translocation. A total of 48 rats received intraperitoneal either Zymosan A (group Z) for systemic inflammation production or sodium chloride solution (controls); they were then further subdivided into three groups of eight animals each, being given, through the tail vein: L-NAME (N-nitro-L-arginine] for inhibition of NO production; SNP (sodium nitroprusside) as NO donor; or sodium chloride as control. After 2 h, the mesenteric lymph node complex was excised, under sterile conditions, and, using standard bacteriological techniques, bacterial translocation was assessed as colony forming units per gram of tissue (CFU/g). Statistical evaluation of the bacteriological data revealed a significant increase of bacterial translocation in all rats subjected to systemic inflammation (group Z) versus controls (P = 0.01) Control rats that were subjected to L-NAME treatment exhibited a statistically significant increase (P = 0.001) in CFU/g compared to sodium chloride treated rats, while SNP treatment revealed no difference in relation to sodium chloride treated rats. Group Z rats, subjected to L-NAME treatment, similarly exhibited a statistically significant increase (P = 0.01) in CFU/g compared to sodium chloride treated rats, while SNP treatment led to a statistical increase of bacterial translocation in relation to sodium chloride treated rats (P = 0.05). The results of this study lead us to suggest that NO appears to participate in the process of bacterial translocation.

The use of the buccal fat pad for reconstruction of oral defects: review of the literature and report of 15 cases.

PURPOSE: Although the buccal fat pad (BFP) was originally used as an alternative method for the closure of small to medium-sized oroantral and oronasal communications, its use has now been extended to use after excision of oral malignancies. This report describes experience with this technique. PATIENTS AND METHODS: The BFP was used as a pedicled graft to reconstruct medium-sized surgical defects of the oral soft and hard tissues in 15 patients suffering from oral malignant tumors. Six of the defects were in the maxilla, 3 in the retromandibular area, and 6 in the cheek and oral commissure. The BFP was left uncovered to epithelialize in 14 patients, and in one it was covered with lyophilized dura. RESULTS: The BFP healed without complications within 3 to 4 weeks in 13 patients, whereas in 2 patients with maxillary defects there was partial loss of the graft, requiring the additional use of an obturator in 1 case and a tongue flap in another to prevent oronasal leakage. Harvesting the graft proved to be extremely easy, and care was only necessary to avoid severing the supporting vascular plexus and the thin capsule covering the BFP. CONCLUSIONS: The findings support the view that the BFP is a useful, easy, and uncomplicated alternative method for the reconstruction of small to medium-sized surgical defects of the oral hard and soft tissues.

Long-term measurement of liver microcirculation in dogs and humans.

BACKGROUND: We verified the possibility of using laser Doppler fluxmetry for prolonged monitoring of hepatic perfusion; we confirm the ability of an implantable laser Doppler microprobe to be in constant 'optical contact' with the liver and thus to transmit a stable microcirculatory signal for a prolonged period of time, and we correlate the response of liver microcirculation to the hepatic artery blood flow reduction in order to estimate this flow by continuous monitoring of microcirculation. METHOD: Hepatic microcirculation was recorded by a single-fiber microprobe implanted in the livers of 8 dogs and of 5 surgical ICU patients. In another 7 dogs, liver microcirculation as well as hepatic artery blood flow were recorded digitally, while an occluder was used to decrease hepatic artery flow. RESULTS: Analysis of the initial data of microcirculation revealed a nonsignificant variation between consecutive time segments, a finding confirming the hypothesis that laser Doppler gives a very stable signal over a long period of time. Polynomial regression analysis, performed on data pairs obtained from microcirculation and hepatic artery blood flow revealed a regression coefficient y = -54.22 + 1.07x + 0.0046x2 (y = hepatic artery blood flow, x = liver microcirculation). CONCLUSION: This finding means that it is possible to watch the hepatic artery flow values continuously by the use of this equation and simple monitoring of liver microcirculation. Thus, the use of laser Doppler fluxmetry with implantable microprobes seems promising as a novel method for continuous assessment of hepatic artery blood flow.

Modulation of intravariceal pressure with pentoxifylline: a possible new approach in the treatment of portal hypertension.

OBJECTIVE: In this study the effect of the hemorheological agent pentoxifylline on the pressure of esophageal varices was investigated in portal hypertensive cirrhotic patients. METHODS: Intravariceal pressure was measured endoscopically using the direct puncture technique in 20 patients. Measurements were obtained under baseline conditions and 30 min after double-blind administration of pentoxifylline (1.4 mg/kg BW, n = 10 patients) or an identical volume of NaCl 0.9% solution (n = 10 patients). RESULTS: Under baseline conditions, intravariceal pressure was similar in pentoxifylline and placebo groups (17.3+/-5.5 mm Hg vs 18.8+/-4.6 mm Hg, respectively; p = N.S.). Placebo administration had no significant effect on intravariceal pressure (18.8+/-4.6 mm Hg vs 18.3+/-4.1 mm Hg; p = N.S.). In contrast, pentoxifylline caused a highly significant reduction of intravariceal pressure, (from 17.3+/-5.5 mm Hg to 11.4+/-5.9 mm Hg; p = 0.0001), the overall mean reduction being 36.1+/-14.1% mm Hg. CONCLUSIONS: We concluded that pentoxifylline, by reducing blood flow viscosity, caused a significant decrease in variceal pressure in patients suffering from portal hypertension.

Blood digestion-induced splanchnic hyperemia and portal blood flow in portal hypertensive rats and the role of octreotide.

We aimed to investigate whether the presence of blood within the intestinal lumen after variceal bleeding would lead to reactive intestinal hyperemia, which in turn could result in the worsening of portal hemodynamics, and thus bleeding recurrence. Two models of portal hypertensive Wistar rats were used: 32 CCl4-cirrhotics with a low index of portal-systemic shunting and 32 that had been previously subjected to portal vein stenosis, with a high index of portal-systemic shunting; 32 Wistar rats served as controls. The rats were divided into four groups, each comprising 8 cirrhotics, 8 portal vein stenosis rats, and 8 controls. Intestinal microcirculation and portal blood flow were assessed by laser-Doppler and transit-time ultrasonic flow probes, respectively, before and 60 min after the injection of 2 ml of blood (groups 1 and 2) or an equal volume of NaCl 0.9% (placebo; groups 3 and 4) into the intestinal lumen. Octreotide (0.2 microg/100 g body weight [BW]) (groups 1 and 3) or NaCl 0.9% (groups 2 and 4) was then given subcutaneously, and 30 min later the final measurements were performed. The presence of blood within the intestinal lumen resulted in an increase in intestinal microcirculation in rats in all groups, while portal blood flow was increased in portal vein stenosis rats and controls, and decreased in cirrhotics. The presence of NaCl 0.9% had no effect. Octreotide, but not NaCl 0.9%, led to a decrease in both intestinal microcirculation and portal blood flow. The findings of this study suggest that intestinal hyperemia induced by digestion of blood in the enteric lumen increases or decreases portal blood flow, the result being strongly related to the portal hypertension model used. Since the main difference between the models was the extent of portal-systemic shunting, this may suggest a relationship between portal blood flow and portal-systemic shunting. This relationship could explain why variceal bleeding stops in some patients but recurs in others.

Influence of enteral nutrition-induced splanchnic hyperemia on the septic origin of splanchnic ischemia.

The purpose of this experimental study was to investigate whether enteral nutrition-induced postprandial intestinal hyperemia has a beneficial effect on the splanchnic ischemia due to sepsis. Fourteen dogs, after exposure to Escherichia coli endotoxin via portal vein administration were grouped according to whether they were fed enterally via a jejunostomy or given a placebo. Systemic hemodynamics; portal vein, hepatic, and superior mesenteric artery blood flow; hepatic and intestinal microcirculation; hepatic tissue PO2; intestinal pHi; and hepatic energy charge were assessed before, during, and after endotoxin infusion as well as during and after enteral or placebo feeding. All splanchnic hemodynamic parameters revealed a statistically significant decline (p = 0.001) during the endotoxin shock period relative to the baseline. After enteral feeding all parameters exhibited a statistically significant increase (p = 0.001) relative to the placebo group. The results of this study led us to suggest that enteral nutrition reverses the lipopolysaccharide infusion-induced splanchnic ischemia.

24-hr measurement of gastric mucosal perfusion in conscious humans.

BACKGROUND/AIMS: Gastric mucosal blood flow estimation in humans is obtained through an endoscope and the time of measurement lasts only a few minutes. Thinking that long-term monitoring of mucosal perfusion would be a significant contribution to the study of gastric physiology, we registered gastric mucosal blood flow continuously for 24 hours, using single fiber laser-Doppler technology. METHODOLOGY: The study was undertaken in 16 healthy subjects (8 of them had their gastric acidity inhibited with a proton pump inhibitor) and in 8 patients with an endoscopically proven, active duodenal ulcer. A 140 cm-long single fiber laser-Doppler microprobe was positioned through a gastrointestinal tube in the middle of the gastric corpus and the mucosal microcirculation was monitored from 14.00 h until 13.59 h the following day. Data were stored and processed to evaluate the probable circadian rhythms, using maximum entropy spectrum analysis. RESULTS: We found that the daily variations of gastric mucosal perfusion follow a circadian rhythm. The respective patterns with maximum and minimum values were: healthy controls, maximum at 02.00, 10.00, 18.00 h and minimum at 5.30, 14.00 and 22.00 h. Healthy controls treated by a proton pump inhibitor, maximum at 02.00, 07.00, 18.00 h and minimum at 04.00, 12.00 and 22.00 h. Ulcer patients, maximum 07.00 and 21.00 h and minimum at 17.00 and 24.00 h. CONCLUSIONS: It is concluded that long-term measurement of gastric mucosal blood flow in conscious humans is feasible and that this factor of gastric physiology follows a concrete circadian rhythm, which is not particularly influenced by acid inhibition, but is completely distorted in ulcer patients.

Hepatic tissue microcirculation, oxygenation and energy charge in ischemia-reperfusion subjected cirrhotic rat liver.

BACKGROUND/AIMS: The hemodynamic disturbances in the cirrhotic liver following severe variceal bleeding and subsequent restoration by blood transfusion is an ischemia/reperfusion injury event which represents the clinical situation of liver dysfunction. Therefore, the aim of this study was to evaluate the microcirculation, oxygenation and energy charge of the cirrhotic rat liver after ischemia/reperfusion. METHODOLOGY: In eight carbon tetrachloride-induced cirrhotic rats and an equal number of controls subjected to 30 minutes of ischemia and 60 minutes of reperfusion by hepatoduodenal ligament clamping, the following parameters were assessed: hepatic microcirculation by laser-Doppler fluxmetry, hepatic tissue oxygenation by a Clark-type electrode, hepatic energy charge by tissue sampling and adenine-nucleotides determination by means of high-performance liquid chromatography. RESULTS: At baseline, liver microcirculation was found to be significantly decreased in the cirrhotics versus controls groups. Ischemia led to a reduction in both groups, while reperfusion improved microcirculation, but not to the baseline level. Oxygenation was reduced during ischemia and restored after reperfusion in both groups. Hepatic energy charge was reduced in the cirrhotics versus controls at baseline, and significantly decreased during ischemia in both groups. At reperfusion, a further reduction was found in the cirrhotic group, while in the control group it was restored to baseline. CONCLUSION: Hepatic microcirculation, oxygenation and energy charge are subjected to different degrees of diminution after ischemia/reperfusion in the cirrhotic rat liver.

Splanchnic ischemia during laparoscopic cholecystectomy.

BACKGROUND: Experimental studies have shown that elevation of intraabdominal pressure by means of gas insufflation produces hemodynamic disturbances in the peritoneal viscera, leading to splanchnic ischemia. The purpose of this clinical investigation is to reproduce the experimental data in humans undergoing laparoscopic cholecystectomy. METHODS: Sixteen females participated in this study. Eight of them (the control group) were subjected to open laparotomy for biliary surgery, while on the remaining eight laparoscopic cholecystectomy was performed. In all patients hepatic microcirculation was registered during the time of operation using the laser-Doppler technique. A single-fiber laser-Doppler microprobe was introduced transcutaneously within the hepatic parenchyma, through a Chiba needle, under direct or laparoscopic vision. Additionally, gastric intramucosal/intramural pH, a low level of which indicates tissue ischemia, was assessed by means of a tonometric nasogastric catheter. Hepatic microcirculation and gastric intramucosal/intramural pH were assessed between controls and pneumoperitoneum-subjected patients, and within the laparoscopic surgery group, i.e., during pneumoperitoneum and after abdominal deflation. RESULTS: Hepatic microcirculation was found to be significantly decreased in laparoscopic surgery patients in relation to controls (22.21 +/- 5.48 vs 57.52 +/- 18.06 perfusion units of flow, P = 0.0001) as was gastric intramural pH (7.15 +/- 0.16 vs 7.37 +/- 0.02, P = 0.003). Similarly, immediately after abdominal deflation, hepatic microcirculation exhibited a sudden elevation (22.21 +/- 5.48 vs 67.49 +/- 7.93 perfusion units of flow, P = 0.0001), while gastric intramural pH return to its normal values (7.15 +/- 0.16 vs 7.43 +/- 0.07, P = 0.0001). CONCLUSIONS: It is concluded that during laparoscopic cholecystectomy abdominal organs are hypoperfused, leading to a splanchnic ischemia environment. The clinical significance of these events remains to be clarified.

Gut ischemia, oxidative stress, and bacterial translocation in elevated abdominal pressure in rats.

The purpose of this experimental study was to investigate whether the increased intraabdominal pressure due to gas insufflation creates intestinal ischemia leading to oxygen free radical production and bacterial translocation. A group of 88 rats were studied, 40 of which were subjected to a 15 mmHg pressure pneumoperitoneum for 60 minutes, with the following parameters being studied: mean arterial pressure after carotid catheterization; intestinal microcirculation by means of the laser-Doppler technique; gut metabolic activity (O2 extraction) by blood sampling from portal vein and carotid artery; intestinal, hepatic, splenic, and lung free radical production (malondialdehyde); and bacterial translocation toward the mesenteric lymph nodes, liver, and spleen at 3 and 18 hours after pneumoperitoneum deflation. The mean arterial pressure exhibited no alterations, whereas the jejunal mucosa microcirculation was significantly decreased (p = 0.0001), as was the gut metabolic activity (p = 0.025). Malondialdehyde was increased in gut mucosa (p = 0.0002), liver (p = 0.02), spleen (p = 0.03), and lung (p = 0.017). Bacterial translocation toward the mesenteric lymph nodes (p = 0.002), spleen (p = 0.002), and liver (p = 0.05) was increased in the 3-hour group; in the 18-hour group bacteria were not found in mesenteric lymph nodes but were in liver (p = 0.008) and spleen (p = 0.035). It is concluded that elevated intraabdominal pressure in the rat leads to intestinal ischemia, oxygen free radical production, and bacterial translocation. These results must be reproduced in humans and their clinical significance clarified.

Increased capillary endothelial leakage in portal hypertensive gastric mucosa: fluorescence microscopy in CCl4-induced cirrhotic rats.

Since portal hypertension affects the gastric mucosa, leading to congestive gastropathy and thus to increased incidence of bleeding, it is one of the possible causes of increased permeability of gastric mucosal capillaries. The aim of this study was the quantitative assessment of the permeability of the gastric mucosal endothelial cells. Eight CCl4-induced cirrhotic rats and eight matched controls were subjected to i.v. injection of FITC-albumin, and a morphometric evaluation of fluorescence in serial histological sections of the gastric mucosa was performed by a video image analysis system. Fluorescence was found to be 0.351 +/- 0.01% of the area scanned in experimental animals versus 0.073 +/- 0.005% in controls, i.e. it was significantly increased by the treatment, which implies a significant endothelial leakage into the extravascular space.