Octreotide in the Control of Post-Sclerotherapy Bleeding from Oesophageal Varices, Ulcers and Oesophagitis

Bleeding from oesophageal varices, oesophageal ulcers or oesophagitis is occasionally massive and difficult to control. Octreotide, a synthetic analogue of somatostin lowers portal pressure and collateral blood flow including that through varices, increases lower oesophageal sphincter pressure, and inhibits the gastric secretion of acid as well as pepsin. Our current experience suggests it is effective in controlling acute variceal haemorrhage. Therefore we have examined the efficacy of octreotide in the control of postsclerotherapy bleeding from oesophageal varices, oesophageal ulcers and oesophagitis. During the study period 77 patients experienced a significant gastrointestinal bleed (blood pressure < 100 mm Hg, pulse > 100 beats per min or the need to transfuse 2 or more units of blood to restore the haemoglobin level) following injection sclerotherapy of oesophageal varices. The source of bleeding was varices in 42 patients, oesophageal ulcers in 31 and oesophagitis in 4. All patients received a continuous intravenous infusion of octreotide (50 μg/h) for between 40–140h. If bleeding was not controlled in the first 12h after commencing octreotide hourly bolus doses (50 μg) for 24h were superimposed on the continuous infusion. Haemorrhage was successfully controlled by an infusion of octreotide in 38 of the 42 patients with bleeding from varices, in 30 of 31 patients with oesophageal ulceration, and all patients with oesophagitis. In the 1 patient with persistent bleeding from oesophageal ulceration and in 2 of the 4 with continued haemorrhage from varices, haemostasis was achieved by hourly boluses of 50 μg octreotide for 24h in addition to the continuous infusion. No major complications were associated with octreotide administration. The results of this study clearly indicate that octreotide is a safe and effective treatment for the control of severe haemorrhage after technically successful injection sclerotherapy.


eeding fol- l
wing injection sclerotherapy of oesophageal varices can occur from 3 sources, the varices themselves, oeso- phageal ulcers and oesophagitis.Persistent or early recurrent bleeding from the varices following injection sclerotherapy is often substantial, and is independ- ently associated with a poorer prognosis1.In contrast, bleeding from oesophageal ulcers or oesophagitis af- ter injection sclerotherapy is usually minor, but can occasionally be massive and difficult to control.We have recently demonstrated that somatostatin is a very effective and safe treatment for the control of recur- rent bleeding following injection sclerotherapy from the varices2, oesophageal ulcers and oesophagitis3.Octreotide, a synthetic analogue of somatostatin, has similar effects as the native hormone on portal pres- sure4, azygos blood flow 5'6 intravariceal pressure7, the gastric secretion of pepsin and acid and on lower oesophageal sphincter pressure9.These properties of somatostatin are thought to be the main mechanisms of action whereby the hormone is effective in control- ling recurrent bleeding from varices, oesophageal ulcers or oesophagitis after sclerotherapy.We have therefore evaluated the efficacy of octreotide in the management of post-sclerotherapy bleeding.

but in deep lesions a venous component could not be excluded.The demographic data, grading ofthe sever- ity according to the criteria of Childs'1, aetiology, blood transfusion requirements, the duration of octreotide therapy and the time interval between the last sclerotherapy and onset of bleeding are shown in Table 1.


PATIENTS AND METHODS

Between January 1989 and December 1992, 225 pa- tients presenting with acute variceal haemorrhage, were treated as an emergency, and 127 patients treated electively on 519 occasions by injection sclerotherapy.Injection sclerotherapy was carried out with a flexible endos

pe using an intravari
eal technique of 2-3 ml aliquots up to a maximum of 20 ml at each session.On some occasions however, the sclerosant was acciden- tally injected paravariceally.If oozing persisted after two 10 min periods of tamponade with the endoscope or overtube, a Sengstaken tube was inserted for 12 h.In patients with acute variceal haemorrhage, injection sclerotherapy was repeated 5-7 days later provided that there was no recurrent bleeding during the inter- vening period.Elective sclerotherapy was then carried out at [3][4] week intervals until the varices were obliter- ated.

During the study period 42 patients experienced significant recurrent bleeding from their varices, (within 24 h following sclerotherapy), 31 from oeso- phageal ulcers and 4 from oesophagitis.A significant haemorrhage was defined as either haematemesis and or melaena accompanied by eithe a systemic distur- bance (systolic blood pressure less than 100 mm Hg and a heart rate greater than 100 beats/min) requiring blood tranfusion to maintain the vital signs, or the necessity to transfuse 2 or more units of blood to restore the haemoglobin levels.To the prehaemorrage levels.In all patients the site of haemorrhage was confirmed endoscopically, apart from those who bled in the first 24 h after sclerotherapy.In these patients the site of haemorrhage was assumed to be from the varices since oesophageal ulceration and oesophagitis after sclerotherapy take longer to develop.The ab- sence of oesophagitis, oesophageal ulceration and other sources of recurrent bleeding determined by repeat endoscopy, carried out after cessation of octreotide therapy, confirmed the diagnosis of recur- rent variceal haemorrhage.In no patient bleeding from oesophageal ulcers or oesophagitis could a variceal component to the haemorrhage be.detected,


Octreotide treatment

Initially, all patients with recurrent bleeding from varices, oesophageal ulcers or oesophagitis were trea- ted with a continuous infusion of 50 gg/h octreotide.If bleeding was not controlled in the first 12 h of treat- ment the patients were given hourly bolus injections of octreotide (50 gg) for 24 h in addition to the continu- ous infusion.


Success of treatment

Successful control of haemorrhage was defined as the cessation of bleeding, as confirmed by the absence of overt signs and the stabilisation of the haemoglobin levels, without the need for blood transfusion.


RESULTS


Recurrent variceal bleeding

Durin

the study period a to
al of 744 courses of sclerotherapy were carried out, 225 to control an acute variceal bleed and 519 electively.Continued or early recurrent bleeding from varices occurred in 42 patients following sclerotherap

nce of 5.6%.The incidence of
recurrent variceal bleeding was much higher after emergency sclerotherapy (40/225; 18%) than following an elective session (2/519; 0.4%).

In the 42 patients who rebled following sclero- therapy the sclerosant had been administered intrava- riceally in 36 and paravariceally in 6 at the session preceding their recurrent haemorrhage.The volume of ethanolamine injected in patients who subsequently rebled (median 10, r nge 3-18 ml) was not signifi- cantly different from that administered to patients who did not rebleed (median 8.5, range 2-20 ml).

A continuous infusion of octreotide controlled per- sistent or recurrent variceal bleeding in 38/42 (90.5%) patients (Table 2).In four patients with severe liver disease (Child's C) bleeding was not controlled during the first 12 h of octreotide infusion as indicated by con inued haematemesis and/or melaena accompanied by a systemic disturbance, and the need to  transfuse blood to maintain the haemoglobin level.In two ofthese patients bleeding was controlled by giving hourly bolus doses of 50 gg octreotide in addition to a continuous infusion.Thus, overall control of bleed- ing was achieved in 40/42 (95%) patients treated with octreotide.The two patients who failed octreotide therapy continued to bleed despite fur- ther sclerotherapy and balloon tamponade, and haemorrhage was eventually controlled by emer- gency oesophageal transection.However, both pa- tients developed hepatorenal syndrome following operation and died.

Recurrent bleeding did not occur in any of the 40 patients after cessation of octreotide therapy and 33 were discharged from hospital.Six patients, all with severe liver disease on admission and all presenting with an acute variceal bleed, died (liver failure n 3, hepatorenal syndrome N 3) in hospital.Thus, the overall hospital mortality was 8/42 (19%).


Oesophageal ulceration

Severe bleeding from oesophageal ulceration occurred in 31 patients, an overall incidence of 4.1% The incidence ofsevere bleeding from oesophageal ulceration was lower after a single course of sclerotherapy to control a variceal bleed (1/225; 0.4%) than after ele- ctive sclerotherapy (30/519; 5.8%).In all pat ents apart from one who bled 29 days after the previous session of sclerotherapy, the varices were patent at the diag- nostic endoscopy, but in no case was a variceal compo- nent to the bleeding observed.However, in all these patients the possibility that there were communicating vessels between the ulcers and either the varices or the complex arrangement of blood vessels in or around the oesophagus cannot be discounted.

In the 31 patients who bled from oesophageal ulcers, the sclerosa t was administered intravariceally in 11, paravariceally in and both intravariceally and paravariceally in 19 at the sclerotherapy session preceeding their haemorrhage.The volume of sclerosant injected in the patients who subsequently bled from their oeso- phageal ulcers (median 9, range 2-20 ml) was not ignificantly different from that injected in patients who did bleed from oesophageal ulcers (median 8, range 2-19 ml).

A continuous infusion of octreotide (50 tg/h) con- trolled appreciable bleeding from oesophageal ulcers in 30/31 (97%) patients.In the one patient in whom bleeding persist

, control of
aemorrhage was achieved by hourly bolus doses of 50 tg octreotide superimposed on their continuous infusion (Table 2).

Bleeding did not recur in any of the patients following cessation of octreotide treatment, and 28 were dis- charged from hospital.However 3 patients, all with severe liver dysfunction (Child's C) and alcoholic liver disease died, one from liver failure and two from the hepatorenal syndrome.


Oesophagitis

Appreciable bleeding from oesophagitis occurred only in 4 patients and in each case followed elective injection sclerotherapy.Thus the incidence of appreciable bleed- ing from oesophagitis after sclerotherapy was 0.54%.In 3 patients, bleeding from oesophagitis after sclerotherapy occurred much later than that from the varices or oesophageal ulcers (Table 1).At endoscopy, 3 of the patients had obliterated varices, and one had a single small patent varix.All patients had undergone numerous courses of injection sclerotherapy (median 7, range 5-14).At the endoscopy prior to presenting with appreciable bleeding from circumferential oesophagitis affecting the distal 5cm of the oesophagus, the sclerosant was injected paravariceally in and both intravariceally and paravariceally in . The amount of ethanolamine injected in the patients who bled from oesophagitis (median 8, range 2-9) was less than in those who did not bleed from oesophagitis (median 10, range 2-20).Early rebleeding after scler- otherapy (10 days) occurred in one patient who had received only pa

variceal ad
inistration of sclero- sant at the endoscopy prior to the development of haemorrhagic oesophagitis.

Octreotide infusion successfully controlled the bleeding in all 4 patients bleeding from oesophagitis (Table 2).No further bleeding occurred after cessation of octreotide treatment and all 4 patients made an uneventful recovery and were discharged home on ranitidine and gaviscon.


DISCUSSION

The incidence of recurrent variceal bleeding following injection sclerotherapy observed in this study is similar to that of previously reported series1.Balloon tamponade is widely used to control recurrent variceal bleeding following sclerotherapy.However, balloon tamponade is associated with serious side effects, in- cluding oesophageal ulceration and perforation in 10-25% ofpatients when used as a stop-gap therapy to control bleeding before definitive therapy2.Following sclerotherapy, the risk of balloon tamponade pro- ducing perforation or ulceration of the oesophagus may be increased, because of the damaging effects of the sclerosant on the oesophageal mucosa13.Consequently, it has been suggested that the period of bal- loon tamponade should be limited to a period of 12 h after sclerotherapy to minimise the risk of local com- plications12.If bleeding persists following a 12 h period of balloon tamponade treatment can be problemati- cal.Further periods of tamponade may control bleeding but the complication rate can become unac- ceptably high.Similarly, a second course of sclero- therapy may control bleeding, but in those patients in which it is not successful, the mortality approaches 100%TM.Furthermore emergency oesophageal tran- section is often unrewarding because of difficulty in mobilising the oesophagus after injection sclerothe- rapy 4 and the high mortality associated with surgical intervention in patients with poor liver function15.We have recently demonstrated that somatostatin is a very effective and safe treatment for the control of recurrent variceal bleeding following sclerotherapy2.The results of the present study in a much larger number of patients suggest that octreotide is equally as effective as somatostatin in controlling post-sclerotherapy variceal bleeding.Furthermore, no complications were associated with octreotide therapy.In 2 of the 4 patients in whom octreotide failed to control haemorrhage, haemostasis was achieved by administering hourly bolus doses of octreotide for 24 h, in addition to the continuous infusion.The ra- tionale for this treatment was based on observations in the portal hypertensive cirrhotic rat which suggested that bolus administrations of somatostatin reduced portal pressure and collateral blood flow to a greater extent than a continuous infusion17'18, observations recently confirmed in man19.Therefore, bolus injec- tions of octreotide together wit a continuous infusion of octreotide, could possibly reduce collateral blood flow to a greater extent than an infusion alone and thus facilitate control of bleeding.This suggestion is supported to some extent in this study, since 2 of the 4 patients stopped bleeding during their treatment with hourly bolus doses of octreotide superimposed on a continuous infusion.

Severe bleeding from oesophageal ulcers or oeso- phagitis following injection sclerotherapy is perhaps more problematical than that of recurrent variceal haemorrhage, since there are fewer treatment options available.Balloon tamponade is contraindicated in these circumstances since pressure necrosis may exacerbate the ulceration or oesophagitis, whilst oeso- phageal transection is difficult and unrewarding in patients with poor liver function15'6.Consequently, effective drug therapy is required to control bleeding from these sources.We have demonstrated that somatostatin is a very effective and safe treatment for the control of severe bleeding from post-sclerotherapy oesophageal ulceration and oesophagitis3.The results of this study indicate that an infusion of 50 lag/h octreotide is equally as effective as somatostatin in these indications, controlling bleed- ing in all 4 patients with oesophagitis and 30/31 patients with oesophageai ulceration.Furthermore, in the patient with oesophageal ulcers who con- tinued to bleed during octreotide infusion, addition of hourly boluses of the somatostatin analogue for 24 controlled haemorrhage.This observation further sup- pors the suggestion that bolus doses of octreotide superimposed on a continuous infusion may be useful in controlling persistent bleeding, possibly by reduc- ing collateral blood flow to a greater extent than an infusion alone.No major side-effects were associated with octreotide administration, with the exception of one patient who experienced a marked increase in the plasma levels of bilirubin and liver enzymes.Following oessation of octreotide administration in this pa-tient the plasma levels of bilirubin and liver enzyme gradually returned to pre-treatment levels.

Although reports on the effects of octreotide on portal pressure have provided conflicting results, the analogue has been consistently demonstrated to lower azygos blood flow5'6.Furthermore, octreotide reduces azygos blood flow to a greater extent than it reduces portal pressures'8.Consequently, the reduction in azy- gos blood flow, indicative of a reduction in collateral blood flow including that through the varices, may be the most important determinant in controlling vari- ceal bleeding.A reduction in collateral blood flow, including that through the complex venous network in the lamina propria of the oesophagus, may also be a major factor in controlling bleeding from post-sclerotherapy ulcers and oesophagitis.In portal hyperten- sion, the 4 layers of veins of the lower oesophagus are grossly dilated, and the large intrinsic vessels which form the varices communicate directly or indirectly with all the remaining venous channels2.It is there- fore impossible to exclude a venous component to the haemorrhage from oesophagitis or oesophageal ulcers after sclerotherapy, since it seems likely that there are communications between these lesions, and other ve- nous components of the oesophagus or peri-oesopha- geal veins.Thus although bleeding could be observed from oesophageal ulcers or from the friable mucosa in oesophagitis and not from the varices, it is not possible to exclude a venous component to the haemorrhage, and a reduction in blood flow through these vessels elicited by octreotide may be an important factor in controlling haemorrhage from these lesions.The pre- cise mechanism whereby octreotide decreases collat- eral blood flow is unknown, but may be mediated directly by eliciting vasoconstriction through specific receptors in these vessels2.Alternatively, octreotide may reduce collateral blood flow via inhibition of the release of vasoactive substances responsible for the hyperdynamic circulation in portal hypertensive patients2.Finally, by increasing lower oesophageal pressure octreotide may decrease blood flow through complex venous network of the lower oesophagus.

Following initial control of haemorrhage, octreo- tide has other effects which may be beneficial in pre- venting recurrent bleeding.Although erosion of the varices by reflux of gastric contents is not thought to contribute to variceal bleedingper se, it may, by disso- lution of the fibrin plug or clot at the site of haemo- stasis, precipitate further bleeding.Octreotide by inhibiting the secretion of gastric acid and pepsin and by increasing lower oesophageal sphincter pressure, may reduce the volume and dige