Early-TIPSS Placement in Patients With Variceal Bleeding
Early-TIPSS Placement in Patients With Variceal Bleeding
One hundred and twenty-eight patients with cirrhosis and variceal bleeding were hospitalised in our ICU between March 2011 and January 2013, and 31 of them fulfilled the inclusion criteria (Figure 1). These patients were matched with 31 patients from our historical cohort for gender, age, MELD score, Child–Pugh score and cardiovascular risks. Clinical and biological patient characteristics in the TIPSS+ or TIPSS− groups are shown in Table 1 and were not different between the 2 groups, except for AST levels (245 ± 260 UI vs. 120 ± 92 UI, P = 0.01) and units of blood transfused in the ICU (4.4 ± 4.5 vs. 2.5 ± 2.5, P = 0.04). Mean follow-up was 7.8 months [0.3–12].
(Enlarge Image)
Figure 1.
Flowchart of patients with cirrhosis hospitalised for variceal bleeding in TIPSS+ group.
TIPSS placement was always technically possible and was performed within the 72 hours after admission in all cases, in a mean 37 ± 30 h; PPG dropped from 18 ± 4.8 mmHg to 7 ± 3.7 mmHg.
The 1-year actuarial probability of remaining free of variceal rebleeding was significantly higher in the TIPSS+ group than in the TIPSS− group (97% vs. 51%, P < 0.001) (Figure 2). Uncontrolled bleeding was observed in 1/31 and 2/31 patients in TIPSS+ and TIPSS− groups respectively (P = 0.55). In the TIPSS+ group, one patient rebled 2 days after TIPSS insertion. TIPSS thrombosis was diagnosed and successfully treated by aspiration of the thrombus and insertion of a second stent. Rebleeding between Day-5 and Week-6 occurred in 0/31 patients in the TIPSS+ group and in 9/31 (29.0%) in TIPS− group (P < 0.001).
(Enlarge Image)
Figure 2.
One-year probability of remaining free of rebleeding in case of variceal bleeding in TIPSS+ and TIPSS− group.
Three patients in the TIPSS− group presented with refractory bleeding requiring salvage TIPSS: two patients died and one had to be transplanted because of deterioration of liver function.
The actuarial 6-week survival rates in the TIPSS+ and TIPSS- groups: (90 ± 7% vs. 84 ± 5% respectively (P = 0.45) and the 1-year survival rates [66.8 ± 9.4% vs. 74.2 ± 7.8% respectively (P = 0.78)] were not different between the two groups (Figure 3). Within 1-year after bleeding a total of 8/31 (25.8%) patients in the TIPSS− group and 9/31 (29.0%) in the TIPSS+ group (P = 0.77) had died. The causes of mortality are shown in Table 2.
(Enlarge Image)
Figure 3.
One-year survival in patients with cirrhosis and variceal bleeding: comparison between TIPSS+ and TIPSS− groups.
One year after bleeding 7/31 patients (22.6%) in the TIPSS+ group and 4/31 (12.9%) in TIPSS− group (P = 0.32) had been transplanted.
The development of HE was similar in the TIPSS+ and TIPSS− groups [14/31 (45.1%) vs. 16/31 (51.6%) patients respectively, (P = 0.61)]. The occurrence of sepsis was not different between the two groups [17/31 (54.8%) vs. 16/31 (51.6%) respectively (P = 0.88)] Table 3.
Acute cardiac failure occurred in 8/31 (25.8%) patients in the TIPSS+ group vs. 2/31 (6.4%) in the TIPSS− group (P = 0.03). The presentation was severe in four cases that were all in the TIPSS+ group, requiring mechanical ventilation (non-invasive ventilation in two patients). In the TIPSS+ group, echocardiography was performed in all patients and was considered normal in 7/8 (87.5%) and revealed dilated cardiomyopathy in one patient who died. Pre-TIPSS HVPG was higher in patients who developed acute cardiac failure compared to patients who did not (22 ± 4 mmHg vs. 17 ± 4 mmHg respectively, P = 0.02). With a cut-off value of 20 mmHg for HVPG before TIPSS placement, the probability of developing acute cardiac failure was 44.4% vs. 5.6%, P = 0.003.
Results
Study Patients
One hundred and twenty-eight patients with cirrhosis and variceal bleeding were hospitalised in our ICU between March 2011 and January 2013, and 31 of them fulfilled the inclusion criteria (Figure 1). These patients were matched with 31 patients from our historical cohort for gender, age, MELD score, Child–Pugh score and cardiovascular risks. Clinical and biological patient characteristics in the TIPSS+ or TIPSS− groups are shown in Table 1 and were not different between the 2 groups, except for AST levels (245 ± 260 UI vs. 120 ± 92 UI, P = 0.01) and units of blood transfused in the ICU (4.4 ± 4.5 vs. 2.5 ± 2.5, P = 0.04). Mean follow-up was 7.8 months [0.3–12].
(Enlarge Image)
Figure 1.
Flowchart of patients with cirrhosis hospitalised for variceal bleeding in TIPSS+ group.
TIPSS placement was always technically possible and was performed within the 72 hours after admission in all cases, in a mean 37 ± 30 h; PPG dropped from 18 ± 4.8 mmHg to 7 ± 3.7 mmHg.
Rebleeding
The 1-year actuarial probability of remaining free of variceal rebleeding was significantly higher in the TIPSS+ group than in the TIPSS− group (97% vs. 51%, P < 0.001) (Figure 2). Uncontrolled bleeding was observed in 1/31 and 2/31 patients in TIPSS+ and TIPSS− groups respectively (P = 0.55). In the TIPSS+ group, one patient rebled 2 days after TIPSS insertion. TIPSS thrombosis was diagnosed and successfully treated by aspiration of the thrombus and insertion of a second stent. Rebleeding between Day-5 and Week-6 occurred in 0/31 patients in the TIPSS+ group and in 9/31 (29.0%) in TIPS− group (P < 0.001).
(Enlarge Image)
Figure 2.
One-year probability of remaining free of rebleeding in case of variceal bleeding in TIPSS+ and TIPSS− group.
Three patients in the TIPSS− group presented with refractory bleeding requiring salvage TIPSS: two patients died and one had to be transplanted because of deterioration of liver function.
Survival
The actuarial 6-week survival rates in the TIPSS+ and TIPSS- groups: (90 ± 7% vs. 84 ± 5% respectively (P = 0.45) and the 1-year survival rates [66.8 ± 9.4% vs. 74.2 ± 7.8% respectively (P = 0.78)] were not different between the two groups (Figure 3). Within 1-year after bleeding a total of 8/31 (25.8%) patients in the TIPSS− group and 9/31 (29.0%) in the TIPSS+ group (P = 0.77) had died. The causes of mortality are shown in Table 2.
(Enlarge Image)
Figure 3.
One-year survival in patients with cirrhosis and variceal bleeding: comparison between TIPSS+ and TIPSS− groups.
One year after bleeding 7/31 patients (22.6%) in the TIPSS+ group and 4/31 (12.9%) in TIPSS− group (P = 0.32) had been transplanted.
Adverse Events (Table 3)
The development of HE was similar in the TIPSS+ and TIPSS− groups [14/31 (45.1%) vs. 16/31 (51.6%) patients respectively, (P = 0.61)]. The occurrence of sepsis was not different between the two groups [17/31 (54.8%) vs. 16/31 (51.6%) respectively (P = 0.88)] Table 3.
Acute cardiac failure occurred in 8/31 (25.8%) patients in the TIPSS+ group vs. 2/31 (6.4%) in the TIPSS− group (P = 0.03). The presentation was severe in four cases that were all in the TIPSS+ group, requiring mechanical ventilation (non-invasive ventilation in two patients). In the TIPSS+ group, echocardiography was performed in all patients and was considered normal in 7/8 (87.5%) and revealed dilated cardiomyopathy in one patient who died. Pre-TIPSS HVPG was higher in patients who developed acute cardiac failure compared to patients who did not (22 ± 4 mmHg vs. 17 ± 4 mmHg respectively, P = 0.02). With a cut-off value of 20 mmHg for HVPG before TIPSS placement, the probability of developing acute cardiac failure was 44.4% vs. 5.6%, P = 0.003.
Source...