Intravenous iloprost bridging to orthotopic liver transplantation in portopulmonary hypertension

Two distinct pulmonary circulatory syndromes have been described in association with advanced liver disease: 1) hepatopulmonary syndrome, and 2) portopulmonary hypertension (PPHTN) 1–3. Hepatopulmonary syndrome is characterised by hypoxaemia caused by pulmonary vascular dilatations and direct arteriovenous shunts (table 1⇓). Most centres consider the syndrome to be an indication for orthotopic liver transplantation (OLT) 2. Complete resolution of the syndrome is well documented 4. PPHTN is a specific condition characterised by an elevated mean pulmonary arterial pressure (P_pa), increased pulmonary vascular resistance (PVR), and a normal pulmonary artery occlusion pressure (PAOP) in the presence of portal hypertension and absence of other potential causes of pulmonary hypertension, such as chronic thromboembolic disease, valvular disease, collagen vascular disease, HIV infection, and ingestion of certain toxins or appetite suppressants. In the World Health Organization (WHO) classification, PPHTN represents a subset of pulmonary arterial hypertension 5. There are only a few case reports of patients with the coexistence of PPHTN and hepatopulmonary syndrome 1. Right-to-left intracardiac shunts through a patent foramen ovale, as a cause of hypoxaemia, should be excluded in this situation.

PPHTN is a well-recognised complication of advanced liver disease, and affects 2–5% of such patients. However, in liver transplant candidates, the prevalence is higher and ranges 4–26%. Many of these patients exhibit high-flow states (normal or low PVR) or increased central blood volume (increased PAOP). Thus, the incidence of true PPHTN (increased PVR) is ∼4–9% 6–8. Prognosis is poor, with a mean survival of 15 months after diagnosis 9. Perioperative mortality in patients who undergo OLT is extremely high, andseems to correlate with the degree of increase in P_pa. Although patients with mild-to-moderate PPHTN (P_pa 25–35 mmHg) can safely undergo transplantation without increased risk 6–8, 10, in patients with a P_pa of 35–49 mmHg, the perioperative mortality rate is ∼50%, and reaches 70–100% in those with a P_pa of >50 mmHg 10. The high perioperative mortality results primarily from right ventricular dysfunction. Thus, severe PPHTN has been considered to be a contraindication to OLT. Nevertheless, thereduction in P_pa and PVR, as well as the improvement in right ventricular function during continuous infusion of epoprostenol, may facilitate successful OLT 11–15. The disadvantage of epoprostenol is its very short half-life and the rapid development of tolerance, necessitating increasing doses of the drug to achieve the same effects. Thus, the more stable prostacyclin analogue iloprost might be an alternative treatment option.

Case report

In August 1996, a 48-yr-old patient underwent laparotomy for a perforated gastric ulcer. Routine laboratory examinations revealed elevated liver enzyme levels and liver cirrhosis as a result of chronic hepatitis C, and chronic alcohol abuse was subsequently diagnosed. After a second episode of variceal bleeding in January 1998, the patient successfully stopped drinking alcohol. The clinical course was complicated by several episodes of ascites and jaundice, and, in December 1999, the patient was evaluated for OLT. At that time, no contraindications were found. Echocardiography revealed a normal rightventricular systolic pressure (RVSP; 27 mmHg), measured over right atrial pressure, and the patient was listed for transplantation. In September 2000, a routine control examination revealed a multifocal hepatocellular carcinoma with three nodules, located in segments IV and VIII. Three transarterial segmental hepatic chemoembolisations were performed between September 2000 and July 2001, and they resulted in partial tumour remission.

In January 2001, the patient was admitted for transplantation when a donor organ became available. Induction of anaesthesia was unremarkable, except for right heart catheterisation, which revealed severe pulmonary hypertension with a P_pa of 50 mmHg. In view of this unexpected major risk factor, the transplantation was cancelled. Further diagnostic work-up excluded diseases of the respiratory system, thromboembolism, heart failure and valvular heart diseases. The alveolar–arterial oxygen tension gradient was 16 mmHg (normal <20 mmHg). Thus, severe PPHTN was diagnosed. Not even retrospectively, reviewing the video sequence of the first echocardiographic examination in December 1999, were there any signs of elevated pulmonary arterial pressure, and the examination performed at the same time showed an RVSP of 74 mmHg and the distinctive pattern of severe pulmonary hypertension.

The patient complained only of slight dyspnoea on exertion (WHO functional class II). Analogous to the protocol of Kuo et al. 16, invasive haemodynamic testing was performed (table 2⇓), and continuous i.v. therapy with iloprost started at a gradually increasing infusion rate of ≤200 µg·day⁻¹ (1.8 ng·kg body weight⁻¹·min⁻¹). Iloprost was chosen since it is routinely used instead of epoprostenol in the present authors' pulmonary hypertension programme, primarily because of its longer half-life, its possible lesser tendency to tolerance development, and the need for lower doses and, therefore, lower costs. After 3 months of iloprost infusion, the patient's 6-min walking distance improved, and a significant reduction in RVSP to <40 mmHg could be achieved.

In August 2001, the patient underwent OLT. The intraoperative course was uneventful. Iloprost was continued throughout the transplant procedure. Cardiac function, monitored by electrocardiography and right heart catheterisation, remained stable and without exacerbation of pulmonary hypertension after reperfusion of the transplanted liver. Low-dose noradrenalin was given for 24 h to stabilise the systemic blood pressure. Immunosuppressive therapy with tacrolimus was commenced. After reconstitution of liver function, the iloprost dose had to be increased to 300 µg·day⁻¹ (2.7 ng·kg body weight⁻¹·min⁻¹) because of an increase in P_pa to 45 mmHg.

Postoperative recovery was favourable and without any major complications. The patient was extubated on postoperative day 1, transferred to the ward on day 3, and discharged from hospital on day 13. Treatment with iloprost was continued. The explanted cirrhotic liver showed a moderately differentiated hepatocellular carcinoma in the left lobe and two circumscribed completely necrotic areas in the right lobe, as well as signs of chronic hepatitis with minimal inflammatory activity. In December 2001, 4 months after liver transplantation, all echocardiographic signs of pulmonary hypertension had disappeared, and the RVSP was 27 mmHg. The patient was completely asymptomatic, and the continuous i.v. therapy with iloprost was stopped. The patient is still doing well 25 months after OLT. Echocardiographic results remain normal. Meanwhile, reinfection with hepatitis C virus has been documented and treatment with pegylated interferon alpha-2a and ribavirin has begun. There are no signs of tumour recurrence.

Discussion

The first patient with severe PPHTN successfully bridged to OLT with continuous i.v. iloprost, a stable prostacyclin analogue, is presented. Pulmonary hypertension resolved 4months after successful transplantation and, subsequently, vasodilator therapy could be stopped. The patient's clinical course >2 yrs after OLT is excellent, and there is no recurrence of pulmonary hypertension.

Until the mid-1990s, severe PPHTN was considered to beanabsolute contraindication to OLT. When the P_pa is >50 mmHg, perioperative mortality is 70–100% 10. In 1997, after encouraging results from the continuous administration of epoprostenol in idiopathic pulmonary arterial hypertension 17, Kuo et al. 18 demonstrated a favourable haemodynamic effect of this treatment in four cases of PPHTN, with a 22–71% decrease in PVR. One year later, the same group reported the first case undergoing successful OLT after 4 months of preoperative treatment with i.v. epoprostenol 11. The patient was taken off epoprostenol 3 months aftertransplantation with near normal pulmonary haemodynamics.

To date, only four other cases who bridged to OLT with continuous i.v. epoprostenol have been described (table 3⇓) 12–15. Although one patient died due to right heart failure 28 days after OLT, the other three are alive and successfully weaned from vasodilator therapy.

The novel aspect of the present case is the treatment with continuous infusion of iloprost instead of epoprostenol. Iloprost is a prostacyclin analogue that has several advantages over epoprostenol. Whereas the latter has to be protected from light and needs to be dissolved in a glycine buffer for i.v. administration, iloprost is much more chemically stable. Moreover, iloprost has a biphasic disposition, with longer half-lives of 3–4 and 30 min, in contrast to the half-life of <5 min of epoprostenol 19. In addition, much lower doses of iloprost are needed for a comparable haemodynamic effect 20.

A recent case series has cast doubt on the efficacy of epoprostenol, since six of the 10 patients died after a mean follow-up time of 9 months, at least two of them possibly due to PPHTN after 5 and 6 months, respectively 12. The same group recently reported a peculiar syndrome of progressive splenomegaly with worsening thrombocytopenia and leukopenia in four patients on continuous i.v. epoprostenol 21. One of them died as a result of sepsis following splenic embolisation and subsequent splenectomy. Thus, the present finding that iloprost has a comparable effect to that of epoprostenol suggests that iloprost might become a valuable treatment alternative in severe PPHTN.

The excessive perioperative mortality in patients with severe PPHTN undergoing OLT suggests that early diagnosis is crucial. Since, as in the present case, patients with even severe PPHTN are not infrequently asymptomatic 10, 22, 23, and PPHTN is a serious but treatable disorder, screening for pulmonary hypertension plays a key role in the management of OLT candidates. For the detection of pulmonary hypertension, ECG and plain chest radiography are unreliable, as their sensitivity is only ∼60% and ∼80%, respectively 23. Transthoracic echocardiography is a sensitive noninvasive diagnostic method. It is considered the screening modality of choice for all forms of pulmonary hypertension, including PPHTN. In a study of 165 OLT candidates 24, thesensitivity and specificity of RVSP at a relatively low cut-off of >30 mmHg (including a right atrial pressure of 5 mmHg), with respect to a P_pa of >25 mmHg, were 100 and96%, respectively. However, half of these patients with invasively confirmed PPHTN had a P_pa of <40 mmHg and could thus undergo OLT without additional treatment. In another study of 74OLT candidates using a higher RVSP cut-off of ≥50 mmHg, sensitivity and specificity were 97 and 72%, respectively 25. Only one patient with an RVSP of<50 mmHg had a P_pa of ≥35 mmHg, but the invasively determined systolic pulmonary arterial pressure was only 55 mmHg. Thus, Doppler echocardiography is an accurate screening test, and all OLT candidates with an RVSP of ≥50 mmHg should undergo right heart catheterisation. Although rare, PPHTN can develop after evaluation for OLT, as was the case in the present patient. In their large series, Colle et al. 24 describe three patients who developed PPHTN within only 3–5 months after the initial evaluation for OLT. In their retrospective review of 43 patients, Krowka et al. 10 found that, in 65%, PPHTN was diagnosed in the operating room immediately prior to OLT. Thus, regular echocardiographic examinations of OLT candidates on the waiting list are necessary, although the optimal screening frequency remains to be determined. Inaccordance with the Mayo Clinic Liver Transplantation Program (M.J. Krowka, Division of Pulmonary and Critical Care, The Mayo Clinic, Rochester, MN, USA, personal communication), the patients on the current authors' waiting list are screened every 12 months if the first echocardiography is normal,and every 6 months if the RVSP is 35–50 mmHg. Patients with an RVSP of >50 mmHg undergo right heart catheterisation and initiation of vasodilator therapy, with theaim of decreasing P_pa to <35–40 mmHg and PVR to<400 dyn·s·cm⁻⁵ with good cardiac output before listing them for OLT.

Since the aetiology of PPHTN is unknown 26, it is difficult to predict whether pulmonary hypertension will resolve after OLT. Literature data indicate that, as in the present case, almost all patients with PPHTN show improved or normalised pulmonary artery pressures within 6 months after OLT 27–30, but this may take up to 2 yrs 31. In literature reviews, it was found that 19 out of 25 (76%) patients ever transplanted for PPHTN showed improved or resolved pulmonary hypertension 6, 8, 10. In individual cases, persistence 6, 14, 32, 33, progression 6, 34, recurrence 35 or even de novo evolution 36 of pulmonary hypertension after OLT have been observed. A comparison of the preoperative haemodynamics revealed that patients who showed improved or resolved pulmonary hypertension had a lower P_pa (43 mmHg; 95% confidence interval (CI) 39–47 mmHg) compared with those who did not (51 mmHg; 95% CI 42–61 mmHg). This difference, however, was not significant (p=0.1). Thus, owing to the large overlap between the two groups, the individual postoperative course cannot be predicted from the severity of PPHTN. It is noteworthy that the present case represents the one with the highest preoperative P_pa of those who showed improved or normalised haemodynamics postoperatively.

In conclusion, it has been shown that iloprost is a valuable alternative for the successful bridging to orthotopic liver transplantation of patients with severe portopulmonary hypertension. Regular echocardiographic screening of orthotopic livertransplantation candidates on the waiting list is mandatory, since portopulmonary hypertension may develop at any time.