André Quinton

Propranolol reduces the rebleeding rate during endoscopic sclerotherapy before variceal obliteration

In patients treated with sclerotherapy, most rebleeding episodes are observed before variceal obliteration. This prospective randomized study aimed to assess if propranolol together with sclerotherapy could reduce the rebleeding rate before variceal obliteration. Seventy-five patients (59 male, 16 female; mean age, 54 +/- 15 years) with cirrhosis (from alcohol abuse in 91%) admitted with upper gastrointestinal bleeding, which was endoscopically proven to originate from ruptured esophageal varices, were included. After initial control of bleeding, the patients were randomized into the following two groups: group 1 treated with sclerotherapy alone (36 patients) and group 2 treated with sclerotherapy plus propranolol (39 patients). They were followed up to variceal obliteration. In group 2, 7 patients rebled as compared with 14 patients treated with sclerotherapy alone (P less than 0.005). When considering only rebleedings from esophageal varices, 4 patients rebled in group 2 vs. 10 in group 1 (P less than 0.10). The total number of rebleeding episodes was lower in group 2 than in group 1 whether considering all causes (8 vs. 17; P less than 0.07) or variceal rebleedings alone (4 vs. 13; P less than 0.01). Mean total blood requirement per patient was lower in group 2 than in group 1 (1.4 +/- 3.4 vs. 2.79 +/- 6.4 units of blood, respectively; P less than 0.01). Mortality was similar in both groups of patients (14% vs. 13% in groups 1 and 2, respectively, NS). It is concluded that patients treated with sclerotherapy should be given propranolol before variceal obliteration.

Agnogenic Myeloid Metaplasia, Portal Hypertension, and Sinusoidal Abnormalities

A patient with agnogenic myeloid metaplasia suffered from gastrointestinal bleeding due to ruptured esophageal varices. The portal vein and its intrahepatic branches were patent. Except for the presence of myeloid cells, mainly megakaryocytes, in the sinusoids, liver histology was more or less normal. However, on Sirius red staining there was marked perisinusoidal fibrosis. In addition to numerous collagen bundles in the Disse space, electron microscopy also revealed the presence of hemopoietic cells, the transformation of perisinusoidal cells into fibroblasticlike or myofibroblasticlike cells, or both, and fragmentary deposits of basement membrane-like material. In the pathogenesis of sinusoidal hypertension as it occurs in agnogenic myeloid metaplasia, all the factors mentioned above should probably be taken into consideration.

Predictors of response to infliximab in luminal Crohn's disease.

AIMS To identify predictive factors of response to infliximab in luminal Crohns disease (CD). PATIENTS AND METHODS All consecutive patients with luminal CD treated with infliximab between October 1999 and March 2003 in Bordeauxs referral centers were included. All had at least 3 months follow-up post infliximab infusion and no prior treatment with infliximab. Response rates were determined 2 and 8 weeks after infusion according to Crohns Disease Activity Index (CDAI) (remission=CDAI<150 and response=CDAI decrease more than 100). RESULTS Among 44 patients (33 female; mean age 35 +/- 14 yr.), 39 (88%) had a clinical response 2 weeks after infusion (79% in remission). At week 8, the rate of response was 61.4% and exclusive colonic involvement predicted sustained response to treatment (P=0.03). The probability of remission at 56 weeks was 21.4%. Multivariate analysis demonstrated that the only factor associated with response duration was initiating immunosuppressive (IS) therapy in women (RR=3.61 95%CI[1.25-10.41], P=0.017). CONCLUSION Exclusive colonic involvement is the only predictive factor of sustained response to infliximab in luminal CD. At the time of infliximab infusion, initiation or modification of IS therapy may favor sustained response, at least in women.

Removal of cellular debris formed in the Disse space in patients with cholestasis.

Using electron microscopy, we investigated how cellular debris, formed in the Disse space during cholestasis, was cleared. Ten patients with cholestasis of varied origin and severity were studied and compared with 10 controls without liver disease. In cholestatic patients, sinusoidal cells contained variable amounts of amylase PAS-positive material. In clean perfusion-fixed sinusoids the endothelial cells often appeared swollen and active, with few fenestrations. Hepatocyte blebs and cellular debris were sometimes seen in the Disse space. Two mechanisms were apparently involved in the clearing process: phagocytosis by macrophages either infiltrated into the Disse space, or forming the barrier; and the passage of debris from the Disse space into the sinusoidal lumen through the endothelial wall. Debris was either forced through enlarged pores or through the wall, with a progressive invagination followed by an outpouching in the lumen. The force, possibly provided by endothelial massage, may not be sufficient to push out cellular debris from the Disse space; morphological data seemed to indicate that endothelial damage may be a necessary factor. Debris present in the lumen was phagocytized by numerous active macrophages. Cellular debris was not observed in the Disse space of control patients.