Lung (2014) 192:421–427 DOI 10.1007/s00408-014-9571-z
INTERSTITIAL LUNG DISEASE
Interstitial Lung Disease in Patients with Hepatopulmonary Syndrome: A Case Series and New Observations S. Shahangian • M. Y. Shino • I. Barjaktarevic I. Susanto • J. A. Belperio • M. C. Fishbein • T. Wang
•
Received: 7 September 2013 / Accepted: 7 March 2014 / Published online: 30 March 2014 Ó Springer Science+Business Media New York 2014
Abstract Hepatopulmonary syndrome (HPS) is characterized by impaired oxygenation due to pulmonary vascular dilatation in patients with end-stage liver disease. At our center, we identified 29 patients who were listed for liver transplantation (LT) with a model for end-stage liver disease exception for HPS between 2001 and 2012. Five of these patients were found to have concurrent interstitial lung disease (ILD). The chest high-resolution computedtomography demonstrated ground-glass opacities and subpleural reticulation, most consistent with nonspecific interstitial pneumonia (NSIP). All four of our patients who underwent LT experienced prolonged hypoxemia postoperatively, with one surgery-related death. However, the three surviving patients had eventual resolution of their hypoxemia with no evidence of ILD progression. In conclusion, we report a high prevalence of ILD, most consistent with NSIP, among patients with HPS. Although there may be increased perioperative risks, the finding of ILD in patients with HPS should not be considered an absolute contraindication to LT.
increased alveolar-arterial oxygen gradient. It is frequently observed among cirrhotic patients awaiting liver transplantation (LT) with an estimated prevalence between 10 and 30 % [1–3]. HPS portends a poor prognosis with a mortality rate more than twice as high as patients with a similar severity of cirrhotic liver disease [4]. The only known effective treatment for HPS is LT. Historically, severe hypoxemia was considered a contraindication to LT, but subsequent studies found significant improvement in oxygenation with no difference in mortality after LT in patients with HPS [5–7]. HPS is now considered an indication for expedited referral and evaluation for LT [8]. The occurrence of interstitial lung disease (ILD) among these patients has not been well described. We have identified a high prevalence of ILD among patients with HPS evaluated for liver transplantation at our center. This case series describes the clinical characteristics, progression of ILD, and outcomes after LT for five patients with HPS and ILD (HPS–ILD).
Keywords Hepatopulmonary syndrome Interstitial lung disease Pulmonary fibrosis Liver transplantation
Method
Introduction Hepatopulmonary syndrome (HPS) is defined by the triad of: liver disease, pulmonary vascular dilatation, and
S. Shahangian (&) M. Y. Shino I. Barjaktarevic I. Susanto J. A. Belperio M. C. Fishbein T. Wang Department of Pulmonary and Critical Care, University of California, Los Angeles, Westwood, USA e-mail: [email protected]
We identified 29 patients at our center with end-stage liver disease who were listed for LT by the United Network for Organ Sharing with a model for end-stage liver disease (MELD) exception for HPS between 2001 and 2012. Nationwide and at our institution, HPS patients with a partial pressure arterial oxygen (PaO2) on room air of less than 60 mmHg qualify for a MELD exception for LT. The presence of HPS was confirmed in all 29 patients with liver cirrhosis based on arterial blood gases showing hypoxemia (PaO2 \ 60 mmHg), and presence of intrapulmonary shunting seen on contrast-enhanced echocardiogram. All patients underwent a chest radiograph, and pulmonary
123
422
function tests. Before 2008, only patients with abnormal chest radiographs underwent a high-resolution computedtomography (HRCT) to assess if there was any underlying interstitial lung disease. However since 2008, HRCT has become a part of the protocol at our institution on all patients diagnosed with HPS. All HRCTs were performed with supine and prone images, with both inspiratory and expiratory cuts, and all scans were performed on the same scanner with 1 mm reconstruction images. The same chest radiologist read and interpreted all scans. Five of the 29 patients listed for LT were found to have evidence of ILD on their chest HRCT.
Case Reports Case 1 A 57-year-old male with cirrhosis secondary to alcohol and hemochromatosis was diagnosed with HPS and ILD 1 year before referral for LT. HRCT showed subpleural reticulation, septal thickening, and mild architectural distortion without honeycombing, most consistent with nonspecific interstitial pneumonia (NSIP) (Fig. 1) (Table 1). Pulmonary function testing (PFT) showed mild restriction with a diffusion capacity for carbon monoxide (DLCO) of 50 % predicted (Table 2). At the time of transplantation 2 years later, the patient became oxygen-dependent and required 6 liters per minute (Lpm) of oxygen supplementation. At this time, the patient had PaO2 of 47 mmHg and FVC at 59 % with a disproportionate decrease in DLCO to 27 % of the predicted value.
Lung (2014) 192:421–427
LT was complicated by postoperative hypoxemia that required prolonged mechanical ventilation (5 days). He also developed urosepsis that required reintubation but was weaned off oxygen and discharged within a month posttransplant. HRCT (Fig. 2) and serial PFTs (Table 2) have continued to show stability of the ILD 2 years after LT.
Case 2 A 65-year-old male with hepatitis C cirrhosis was diagnosed with HPS and ILD 4 years before referral for LT. HRCT showed an NSIP pattern with mild restriction on spirometry (Table 1). Interestingly, during the 18 months awaiting LT, the patient had improvement of his hypoxemia, with resolution of his HPS confirmed by contrast echocardiogram. However, he developed pulmonary hypertension with a right-ventricular systolic pressure of 71 mmHg on transthoracic echocardiogram. A right heart catheterization confirmed a mean pulmonary artery pressure of 40 mmHg. He was placed on epoprostenol with improvement in his pulmonary artery pressure. At the time of transplantation, he required 2 Lpm of oxygen with a PaO2 of 78 mmHg and FVC at 54 % and DLCO 37 % of the predicted value (Table 2). LT was complicated by postsurgical bleeding that required reexploration, but he was successfully extubated on postoperative day (POD) 3. His oxygen and epoprostenol were weaned off, and he was discharged home on room air with sildenafil. HRCT and PFTs have not shown any ILD progression in more than 2 years of follow-up.
Fig. 1 HRCT for case 1 before LT
A
C
123
B HRCT performed 1 month prior to liver transplantation revealed a moderate interstitial pattern with septal thickening and mild architectural distortion in a subpleural distribution predominantly in the bilateral lower lobes and in the left upper lobe. There was no evidence of honeycombing or bronchiectasis with a pulmonary fibrosis pattern most consistent with fibrotic NSIP. Images A, B, and C, represent the upper, middle and lower lung fields respectively.
Lung (2014) 192:421–427
423
Table 1 Patient demographics Case
1
2
3
4
5
Age (years)
57
65
66
48
41
Sex
Male
Male
Male
Female
Female
Race
Caucasian
Caucasian
Hispanic
Hispanic
Caucasian
Etiology
Alcohol and hemochromatosis
Hepatitis C
Alcohol
Hepatitis C
Cryptogenic
Previous history of smoking
No
No
No
Yes
No
Home oxygen before OLT (L)
6
2
5
14
8–15
Physiologic MELD
10
28
12
14
8
MELD exception score
34
29
29
Unknown
32
CHILD Pugh score
A
B
B
Unknown
B
HRCT pattern
Peripheral reticulation, architectural and pleural parenchymal interface distortion
Peripheral reticulation, architectural and pleural parenchymal interface distortion
Minimal peripheral reticular prominence diffusely, and ground glass opacification in bilateral lower lobes
Diffuse subpleural reticulation, architectural and pleural parenchymal interface distortion
Intralobular septal thickening with traction bronchiectasis and architectural distortion without significant honeycombing
PFT pattern
Restrictive and diffusion impairment
Restrictive and diffusion impairment
Diffusion impairment
Diffusion impairment
Restrictive and diffusion impairment
Duration of hospitalization
23
18
15
NA
NA
Postoperative complication
Hypoxemia/ reintubation
Bleeding/ hypoxemia
Hypoxemia/reintubation
NA
Air embolus/refractory hypoxemia/death
Days intubated
5?3
3
4?5
NA
NA
O2 requirement at discharge
No
No
4L
NA
NA
Alive
Yes
Yes
Yes
No
No
Survival posttransplantation
Alive at 21 months
Alive at 23 months
Alive at 32 months
Passed away 8 years after first LT
Died 7 days after aborted OLT
Table 2 Pulmonary function tests (before and after transplantation) Case
Initial parametersb Months to LT
FVC
Last parameters pretransplantc DLCO
Room air PaO2
TLC
FVC
Last parameters posttransplantd
DLCO
Room air PaO2
A-a gradient
Months from LT
FVC
DLCO
1 2
31 18
78 87
50 41
109 56
66 77
59 81
27 37
47 78
61 51
10 6
63 81
43 59
3
17
82
30
58
102
84
28
52
57
NA
NA
NA
4
NA
105
47
76
NA
103
41
NA
38
NA
NA
NA
5
30
76
30
63
72
73
25
49
53
NA
NA
NA
a
LT liver transplantation; FVC forced vital capacity; DLCO diffusing capacity; PaO2 partial pressure of oxygen; RA room air; A-a alveolar arterial a
All values for Case 4 were obtained after the first liver transplantation while awaiting a second transplant
b
First set of parameters (FVC, DLCO, or PaO2) that were available before liver transplantation
c
Last set of parameters before transplantation. Arterial blood gases were obtained within 1 month of surgery. Pulmonary function tests were obtained 2–12 months before surgery
d
Last set of parameters that were available posttransplant
123
424
Lung (2014) 192:421–427
Fig. 2 HRCT for case 1, post LT
A
B HRCT performed two years after liver transplantation remained largely unchanged with lower greater than upper lobe subpleural reticulation and architectural and pleural parenchymal interface distortion. Images A, B, and C, represent the upper, middle and lower lung fields respectively.
C Case 3 A 66-year-old male with alcoholic cirrhosis was diagnosed with HPS and ILD during his LT evaluation. HRCT showed mild ILD in an NSIP pattern (Table 1). At the time of transplantation 1 year later, he required 5 Lpm of oxygen with a PaO2 of 52 mmHg and FVC at 84 % and DLCO 28 % of the predicted value (Table 2). LT was complicated by severe postoperative hypoxemia. He was extubated on POD 4 but was reintubated on POD 23 for hypoxemic respiratory failure thought to be due to persistent HPS. On POD 28, he was successfully extubated and subsequently discharged to a skilled nursing facility on POD 34 with 3–4 Lpm of oxygen. At his last follow-up 4 months after LT, he had only mild functional limitation, requiring oxygen only during exertion.
Case 4 A 48-year-old female with hepatitis C cirrhosis with hepatocellular carcinoma was referred to our center 6 years after LT at another center. She required oxygen for the 2 months before her transplant due to HPS. She was weaned off oxygen with no functional limitation but later developed recurrent hepatitis C cirrhosis. At her initial evaluation at our center, contrast echocardiography demonstrated an intrapulmonary shunt consistent with recurrent HPS. HRCT showed a moderate NSIP pattern with a DLCO of 47 %. She developed
123
posttransplant lymphoproliferative disorder with liver involvement and underwent chemotherapy. She eventually expired secondary to sepsis and multisystem organ failure. With regards to her ILD, HRCT and PFTs remained stable over a 2-year period of follow-up. Case 5 A 41-year-old female with cryptogenic cirrhosis diagnosed 20 years earlier was referred for LT. She had a longstanding history of HPS and mild interstitial lung disease (Table 1). HRCT showed mild NSIP (Fig. 3), and PFTs showed mild restriction with a disproportionate decrease in DLCO to 30 % predicted. During the next 2 years until LT, there was no evidence of ILD progression based on HRCT and serial spirometry. However, she continued to have severe hypoxemia, which was attributed to progressive HPS. At the time of her LT, she required 8–15 L of oxygen at rest with a PaO2 of 49 mmHg. The FVC was at 73 % and the DLCO was 25 % of the predicted value at the time of transplant (Table 2). Her transplant was complicated by acute hypotension and hypoxemia due to an air embolus, confirmed by an intraoperative transesophageal echocardiogram. The surgery was aborted and she expired 1 week later from hypoxemic respiratory failure. A postmortem autopsy of the lung revealed moderate intimal thickening of small- and medium-sized arteries consistent with HPS as well as subpleural thickening, honeycombing, and fibroblastic foci consistent with a usual interstitial pneumonia pattern (UIP) (Fig. 4).
Lung (2014) 192:421–427
425
Fig. 3 HRCT for case 5 before LT
A
B HRCT performed1-month prior to the attempted liver transplantation revealed bilateral lower lobe greater than upper lobe interlobular and intralobular septal thickening with traction bronchiectasis and bronchiolectasis, loss of pleuroparenchymal interface, and architectural distortion without significant honeycombing. Images A, B, and C, represent the upper, middle and lower lung fields respectively.
C Fig. 4 Post-mortem Autopsy for case 5
123
426
Discussion HPS is a progressive pulmonary vascular disease with a median survival of less than 1 year [3], and LT remains the only effective treatment. Although HPS is traditionally defined as hypoxemia related to pulmonary vascular disease in the absence of significant pulmonary disease, patients often have concurrent pulmonary disease, and at our center, we have identified a high prevalence (17 %) of ILD in the 29 HPS patients listed for LT with a MELD exception. These patients were able to obtain a MELD exception, because their severe hypoxemia was judged to be predominantly related to their HPS rather than their concurrent pulmonary disease. To our knowledge, this is the first case series focusing on ILD in patients with HPS undergoing LT. PFTs showed mostly mild restriction with a disproportionate decrease in the DLCO. HRCT showed lower lobe predominant subpleural reticulation, septal thickening, and mild architectural distortion without significant traction bronchiectasis or honeycombing, overall consistent with an NSIP pattern. However, the one postmortem histopathologic sample available from this series demonstrated subpleural fibrosis, honeycomb change, and fibroblastic foci, consistent with UIP. A few prior studies have noted concurrent ILD in patients with HPS. A retrospective study conducted by Gupta et al. from a tertiary care center found mild ILD in 4 (12 %) of 33 patients with HPS [5]. Another multicenter study of 218 patients undergoing evaluation for LT reported a higher incidence of ILD in patients with HPS: 7 versus 1 % in patients without HPS (p = 0.02) based on chest radiography [4]. The relationship between HPS and pulmonary fibrosis has yet to be elucidated. In our series, ILD was identified before initiation of oxygen, making the toxic effect of longterm oxygen therapy an unlikely etiology. Both cirrhosis and pulmonary fibrosis are fibrotic disorders characterized by epithelial/endothelial injury with dysregulated healing leading to fibroproliferation and tissue remodeling. Although the initial insult is different for these two organs, there may be considerable overlap in the pathogenesis of fibrosis [9]. The pathogenic hallmark of HPS is pulmonary precapillary and capillary dilation, which may be due to the decreased hepatic clearance of inflammatory and angiogenic cytokines. Animal models of HPS have demonstrated the involvement of several cytokines/growth factors, such as TNF-a, IL-1, TGF-b, endothelin-1, and VEGF [10, 11]. Importantly, these same cytokines/growth factors appear to be involved in the pathogenesis of ILDs [12–14]. There also may be factors among HPS patients that predispose them to developing lung injury. For example, HPS patients may have an increased incidence of chemical pneumonitis from
123
Lung (2014) 192:421–427
gastroesophageal reflux and aspiration due to increased intra-abdominal pressures or a tendency to sleep flat due to platypnea. Further observational and translational studies are required to better delineate the true link between these two clinical entities. Several studies have reported increased perioperative complications and mortality in patients with HPS, particularly in those with advanced HPS [7, 15–18]. One retrospective study found that 11 (57 %) of 23 patients with HPS had gas exchange impairment after LT [7]. Two of these patients had severe refractory hypoxemia and died from multiorgan failure. The remaining nine patients were weaned off oxygen over a median duration of 3 months [7]. Another study found that 5 (24 %) of 21 patients with HPS had hypoxemic respiratory failure that required prolonged mechanical ventilation posttransplant (median duration 18 days) [16]. Four (80 %) of these patients were weaned off oxygen at a median of 66 days. The study reported a 7 % 1-year mortality among patients with HPS and 14 % mortality among patients with severe HPS (PaO2 \ 50 mmHg) [16]. The outcomes in our HPS-ILD case series appear to be comparable to outcomes in these previously described HPS case series. All four patients with HPS and ILD who received LT at our center had perioperative complications. The three patients who received successful LT had postoperative hypoxemia that required prolonged mechanical ventilation. However, most were weaned off oxygen with no significant long-term sequelae. One patient developed an intraoperative air embolus, with resultant severe hypoxic respiratory failure and death. Our case series has a number of limitations given that it is a case series with a small sample size with all patients selected from one liver transplant center. Two of the patients identified as having ILD had no evidence of a significant restrictive pattern on PFT but were identified as having ILD based on evidence of clinically significant fibrosis on their HRCT. In addition, before 2008, patients underwent HRCT of the chest only if they had abnormalities on their chest radiograph, and it was not until 2008 that HRCT became part of the protocol on all patients with HPS. Thus, early pulmonary fibrosis could have been missed in some patients in the earlier era, resulting in an underestimation in the prevalence of ILD between 2000 and 2007. Lastly, this case series does not include those patients that were rejected from being placed on the LT list given the extent of their pulmonary fibrosis, creating an inherent selection bias in who is included in our case series. Despite these limitations, we did identify a high prevalence of ILD among patients with HPS who were listed for LT at our center with a MELD exception. The ILD in these patients has the radiographic appearance of NSIP, but our one histopathologic sample demonstrated features
Lung (2014) 192:421–427
consistent with UIP. There were perioperative complications with LT in all of these patients, but the surviving patients had eventual resolution of their hypoxemia with no evidence of ILD progression. The coexistence of HPS and ILD raises difficult questions regarding the appropriateness and timing of listing for liver transplantation. We believe that the presence of ILD should not be an absolute contraindication to LT in patients with HPS. However, when possible, patients with HPS-ILD should be monitored for ILD stability before LT. Conflict of interest
None.
References 1. Arguedas MR, Singh H, Faulk DK, Fallon MB (2007) Utility of pulse oximetry screening for hepatopulmonary syndrome. Clin Gastroenterol Hepatol 5(6):749–754 2. Lima BL, Franca AV, Pazin-Filho A et al (2004) Frequency, clinical characteristics, and respiratory parameters of hepatopulmonary syndrome. Mayo Clin Proc 79(1):42–48 3. Schenk P, Schoniger-Hekele M, Fuhrmann V, Madl C, Silberhumer G, Muller C (2003) Prognostic significance of the hepatopulmonary syndrome in patients with cirrhosis. Gastroenterology 125(4):1042–1052 4. Fallon MB, Krowka MJ, Brown RS et al (2008) Impact of hepatopulmonary syndrome on quality of life and survival in liver transplant candidates. Gastroenterology 135(4):1168–1175 5. Collisson EA, Nourmand H, Fraiman MH et al (2002) Retrospective analysis of the results of liver transplantation for adults with severe hepatopulmonary syndrome. Liver Transpl 8(10): 925–931 6. Deberaldini M, Arcanjo AB, Melo E et al (2008) Hepatopulmonary syndrome: morbidity and survival after liver transplantation. Transpl Proc 40(10):3512–3516
427 7. Taille C, Cadranel J, Bellocq A et al (2003) Liver transplantation for hepatopulmonary syndrome: a ten-year experience in Paris, France. Transplantation 75(9):1482–1489 discussion 1446-1487 8. Murray KF, Carithers RL Jr (2005) AASLD practice guidelines: evaluation of the patient for liver transplantation. Hepatology 41(6):1407–1432 9. Mercer PF, Chambers RC (2013) Coagulation and anticoagulation signaling in fibrosis. Biochim Biophys Acta 1832:1018–1027 10. Liu L, Liu N, Zhao Z et al (2012) TNF-a neutralization improves experimental hepatopulmonary syndrome in rats. Liver Int 32(6):1018–1026 11. Zhang J, Luo B, Tang L et al (2009) Pulmonary angiogenesis in a rat model of hepatopulmonary syndrome. Gastroenterology 136(3):1070–1080 12. Keane M, Strieter R, Lynch P, Belperio J (2006) Inflammation and angiogenesis in fibrotic lung disease. Sem Respir Crit Care Med 27(6):589–599 13. Kapanci Y, Desmouliere A, Pache JC, Redard M, Redard M, Gabbiani G (1995) Cytoskeletal protein modulation in pulmonary alveolar myofibroblasts during idiopathic pulmonary fibrosis. Possible role of transforming growth factor beta and tumor necrosis factor alpha. Am J Respir Crit Care Med 152:2163–2169 14. Michel RP, Stewart DJ, Sheppard M, Hamid Q, Corrin B (1993) Expression of endothelin-1 in lungs of patients with cryptogenic fibrosing alveolitis. Lancet 341(8860):1550–1554 15. Swanson KL, Wiesner RH, Krowka MJ (2005) Natural history of hepatopulmonary syndrome: Impact of liver transplantation. Hepatology 41(5):1122–1129 16. Gupta S, Castel H, Rao RV et al (2010) Improved survival after liver transplantation in patients with hepatopulmonary syndrome. Am J Transpl 10(2):354–363 17. Krowka MJ, Mandell MS, Ramsay MA et al (2004) Hepatopulmonary syndrome and portopulmonary hypertension: a report of the multicenter liver transplant database. Liver Transpl 10(2): 174–182 18. Arguedas MR, Abrams GA, Krowka MJ, Fallon MB (2003) Prospective evaluation of outcomes and predictors of mortality in patients with hepatopulmonary syndrome undergoing liver transplantation. Hepatology 37(1):192–197
123
INTERSTITIAL LUNG DISEASE
Interstitial Lung Disease in Patients with Hepatopulmonary Syndrome: A Case Series and New Observations S. Shahangian • M. Y. Shino • I. Barjaktarevic I. Susanto • J. A. Belperio • M. C. Fishbein • T. Wang
•
Received: 7 September 2013 / Accepted: 7 March 2014 / Published online: 30 March 2014 Ó Springer Science+Business Media New York 2014
Abstract Hepatopulmonary syndrome (HPS) is characterized by impaired oxygenation due to pulmonary vascular dilatation in patients with end-stage liver disease. At our center, we identified 29 patients who were listed for liver transplantation (LT) with a model for end-stage liver disease exception for HPS between 2001 and 2012. Five of these patients were found to have concurrent interstitial lung disease (ILD). The chest high-resolution computedtomography demonstrated ground-glass opacities and subpleural reticulation, most consistent with nonspecific interstitial pneumonia (NSIP). All four of our patients who underwent LT experienced prolonged hypoxemia postoperatively, with one surgery-related death. However, the three surviving patients had eventual resolution of their hypoxemia with no evidence of ILD progression. In conclusion, we report a high prevalence of ILD, most consistent with NSIP, among patients with HPS. Although there may be increased perioperative risks, the finding of ILD in patients with HPS should not be considered an absolute contraindication to LT.
increased alveolar-arterial oxygen gradient. It is frequently observed among cirrhotic patients awaiting liver transplantation (LT) with an estimated prevalence between 10 and 30 % [1–3]. HPS portends a poor prognosis with a mortality rate more than twice as high as patients with a similar severity of cirrhotic liver disease [4]. The only known effective treatment for HPS is LT. Historically, severe hypoxemia was considered a contraindication to LT, but subsequent studies found significant improvement in oxygenation with no difference in mortality after LT in patients with HPS [5–7]. HPS is now considered an indication for expedited referral and evaluation for LT [8]. The occurrence of interstitial lung disease (ILD) among these patients has not been well described. We have identified a high prevalence of ILD among patients with HPS evaluated for liver transplantation at our center. This case series describes the clinical characteristics, progression of ILD, and outcomes after LT for five patients with HPS and ILD (HPS–ILD).
Keywords Hepatopulmonary syndrome Interstitial lung disease Pulmonary fibrosis Liver transplantation
Method
Introduction Hepatopulmonary syndrome (HPS) is defined by the triad of: liver disease, pulmonary vascular dilatation, and
S. Shahangian (&) M. Y. Shino I. Barjaktarevic I. Susanto J. A. Belperio M. C. Fishbein T. Wang Department of Pulmonary and Critical Care, University of California, Los Angeles, Westwood, USA e-mail: [email protected]
We identified 29 patients at our center with end-stage liver disease who were listed for LT by the United Network for Organ Sharing with a model for end-stage liver disease (MELD) exception for HPS between 2001 and 2012. Nationwide and at our institution, HPS patients with a partial pressure arterial oxygen (PaO2) on room air of less than 60 mmHg qualify for a MELD exception for LT. The presence of HPS was confirmed in all 29 patients with liver cirrhosis based on arterial blood gases showing hypoxemia (PaO2 \ 60 mmHg), and presence of intrapulmonary shunting seen on contrast-enhanced echocardiogram. All patients underwent a chest radiograph, and pulmonary
123
422
function tests. Before 2008, only patients with abnormal chest radiographs underwent a high-resolution computedtomography (HRCT) to assess if there was any underlying interstitial lung disease. However since 2008, HRCT has become a part of the protocol at our institution on all patients diagnosed with HPS. All HRCTs were performed with supine and prone images, with both inspiratory and expiratory cuts, and all scans were performed on the same scanner with 1 mm reconstruction images. The same chest radiologist read and interpreted all scans. Five of the 29 patients listed for LT were found to have evidence of ILD on their chest HRCT.
Case Reports Case 1 A 57-year-old male with cirrhosis secondary to alcohol and hemochromatosis was diagnosed with HPS and ILD 1 year before referral for LT. HRCT showed subpleural reticulation, septal thickening, and mild architectural distortion without honeycombing, most consistent with nonspecific interstitial pneumonia (NSIP) (Fig. 1) (Table 1). Pulmonary function testing (PFT) showed mild restriction with a diffusion capacity for carbon monoxide (DLCO) of 50 % predicted (Table 2). At the time of transplantation 2 years later, the patient became oxygen-dependent and required 6 liters per minute (Lpm) of oxygen supplementation. At this time, the patient had PaO2 of 47 mmHg and FVC at 59 % with a disproportionate decrease in DLCO to 27 % of the predicted value.
Lung (2014) 192:421–427
LT was complicated by postoperative hypoxemia that required prolonged mechanical ventilation (5 days). He also developed urosepsis that required reintubation but was weaned off oxygen and discharged within a month posttransplant. HRCT (Fig. 2) and serial PFTs (Table 2) have continued to show stability of the ILD 2 years after LT.
Case 2 A 65-year-old male with hepatitis C cirrhosis was diagnosed with HPS and ILD 4 years before referral for LT. HRCT showed an NSIP pattern with mild restriction on spirometry (Table 1). Interestingly, during the 18 months awaiting LT, the patient had improvement of his hypoxemia, with resolution of his HPS confirmed by contrast echocardiogram. However, he developed pulmonary hypertension with a right-ventricular systolic pressure of 71 mmHg on transthoracic echocardiogram. A right heart catheterization confirmed a mean pulmonary artery pressure of 40 mmHg. He was placed on epoprostenol with improvement in his pulmonary artery pressure. At the time of transplantation, he required 2 Lpm of oxygen with a PaO2 of 78 mmHg and FVC at 54 % and DLCO 37 % of the predicted value (Table 2). LT was complicated by postsurgical bleeding that required reexploration, but he was successfully extubated on postoperative day (POD) 3. His oxygen and epoprostenol were weaned off, and he was discharged home on room air with sildenafil. HRCT and PFTs have not shown any ILD progression in more than 2 years of follow-up.
Fig. 1 HRCT for case 1 before LT
A
C
123
B HRCT performed 1 month prior to liver transplantation revealed a moderate interstitial pattern with septal thickening and mild architectural distortion in a subpleural distribution predominantly in the bilateral lower lobes and in the left upper lobe. There was no evidence of honeycombing or bronchiectasis with a pulmonary fibrosis pattern most consistent with fibrotic NSIP. Images A, B, and C, represent the upper, middle and lower lung fields respectively.
Lung (2014) 192:421–427
423
Table 1 Patient demographics Case
1
2
3
4
5
Age (years)
57
65
66
48
41
Sex
Male
Male
Male
Female
Female
Race
Caucasian
Caucasian
Hispanic
Hispanic
Caucasian
Etiology
Alcohol and hemochromatosis
Hepatitis C
Alcohol
Hepatitis C
Cryptogenic
Previous history of smoking
No
No
No
Yes
No
Home oxygen before OLT (L)
6
2
5
14
8–15
Physiologic MELD
10
28
12
14
8
MELD exception score
34
29
29
Unknown
32
CHILD Pugh score
A
B
B
Unknown
B
HRCT pattern
Peripheral reticulation, architectural and pleural parenchymal interface distortion
Peripheral reticulation, architectural and pleural parenchymal interface distortion
Minimal peripheral reticular prominence diffusely, and ground glass opacification in bilateral lower lobes
Diffuse subpleural reticulation, architectural and pleural parenchymal interface distortion
Intralobular septal thickening with traction bronchiectasis and architectural distortion without significant honeycombing
PFT pattern
Restrictive and diffusion impairment
Restrictive and diffusion impairment
Diffusion impairment
Diffusion impairment
Restrictive and diffusion impairment
Duration of hospitalization
23
18
15
NA
NA
Postoperative complication
Hypoxemia/ reintubation
Bleeding/ hypoxemia
Hypoxemia/reintubation
NA
Air embolus/refractory hypoxemia/death
Days intubated
5?3
3
4?5
NA
NA
O2 requirement at discharge
No
No
4L
NA
NA
Alive
Yes
Yes
Yes
No
No
Survival posttransplantation
Alive at 21 months
Alive at 23 months
Alive at 32 months
Passed away 8 years after first LT
Died 7 days after aborted OLT
Table 2 Pulmonary function tests (before and after transplantation) Case
Initial parametersb Months to LT
FVC
Last parameters pretransplantc DLCO
Room air PaO2
TLC
FVC
Last parameters posttransplantd
DLCO
Room air PaO2
A-a gradient
Months from LT
FVC
DLCO
1 2
31 18
78 87
50 41
109 56
66 77
59 81
27 37
47 78
61 51
10 6
63 81
43 59
3
17
82
30
58
102
84
28
52
57
NA
NA
NA
4
NA
105
47
76
NA
103
41
NA
38
NA
NA
NA
5
30
76
30
63
72
73
25
49
53
NA
NA
NA
a
LT liver transplantation; FVC forced vital capacity; DLCO diffusing capacity; PaO2 partial pressure of oxygen; RA room air; A-a alveolar arterial a
All values for Case 4 were obtained after the first liver transplantation while awaiting a second transplant
b
First set of parameters (FVC, DLCO, or PaO2) that were available before liver transplantation
c
Last set of parameters before transplantation. Arterial blood gases were obtained within 1 month of surgery. Pulmonary function tests were obtained 2–12 months before surgery
d
Last set of parameters that were available posttransplant
123
424
Lung (2014) 192:421–427
Fig. 2 HRCT for case 1, post LT
A
B HRCT performed two years after liver transplantation remained largely unchanged with lower greater than upper lobe subpleural reticulation and architectural and pleural parenchymal interface distortion. Images A, B, and C, represent the upper, middle and lower lung fields respectively.
C Case 3 A 66-year-old male with alcoholic cirrhosis was diagnosed with HPS and ILD during his LT evaluation. HRCT showed mild ILD in an NSIP pattern (Table 1). At the time of transplantation 1 year later, he required 5 Lpm of oxygen with a PaO2 of 52 mmHg and FVC at 84 % and DLCO 28 % of the predicted value (Table 2). LT was complicated by severe postoperative hypoxemia. He was extubated on POD 4 but was reintubated on POD 23 for hypoxemic respiratory failure thought to be due to persistent HPS. On POD 28, he was successfully extubated and subsequently discharged to a skilled nursing facility on POD 34 with 3–4 Lpm of oxygen. At his last follow-up 4 months after LT, he had only mild functional limitation, requiring oxygen only during exertion.
Case 4 A 48-year-old female with hepatitis C cirrhosis with hepatocellular carcinoma was referred to our center 6 years after LT at another center. She required oxygen for the 2 months before her transplant due to HPS. She was weaned off oxygen with no functional limitation but later developed recurrent hepatitis C cirrhosis. At her initial evaluation at our center, contrast echocardiography demonstrated an intrapulmonary shunt consistent with recurrent HPS. HRCT showed a moderate NSIP pattern with a DLCO of 47 %. She developed
123
posttransplant lymphoproliferative disorder with liver involvement and underwent chemotherapy. She eventually expired secondary to sepsis and multisystem organ failure. With regards to her ILD, HRCT and PFTs remained stable over a 2-year period of follow-up. Case 5 A 41-year-old female with cryptogenic cirrhosis diagnosed 20 years earlier was referred for LT. She had a longstanding history of HPS and mild interstitial lung disease (Table 1). HRCT showed mild NSIP (Fig. 3), and PFTs showed mild restriction with a disproportionate decrease in DLCO to 30 % predicted. During the next 2 years until LT, there was no evidence of ILD progression based on HRCT and serial spirometry. However, she continued to have severe hypoxemia, which was attributed to progressive HPS. At the time of her LT, she required 8–15 L of oxygen at rest with a PaO2 of 49 mmHg. The FVC was at 73 % and the DLCO was 25 % of the predicted value at the time of transplant (Table 2). Her transplant was complicated by acute hypotension and hypoxemia due to an air embolus, confirmed by an intraoperative transesophageal echocardiogram. The surgery was aborted and she expired 1 week later from hypoxemic respiratory failure. A postmortem autopsy of the lung revealed moderate intimal thickening of small- and medium-sized arteries consistent with HPS as well as subpleural thickening, honeycombing, and fibroblastic foci consistent with a usual interstitial pneumonia pattern (UIP) (Fig. 4).
Lung (2014) 192:421–427
425
Fig. 3 HRCT for case 5 before LT
A
B HRCT performed1-month prior to the attempted liver transplantation revealed bilateral lower lobe greater than upper lobe interlobular and intralobular septal thickening with traction bronchiectasis and bronchiolectasis, loss of pleuroparenchymal interface, and architectural distortion without significant honeycombing. Images A, B, and C, represent the upper, middle and lower lung fields respectively.
C Fig. 4 Post-mortem Autopsy for case 5
123
426
Discussion HPS is a progressive pulmonary vascular disease with a median survival of less than 1 year [3], and LT remains the only effective treatment. Although HPS is traditionally defined as hypoxemia related to pulmonary vascular disease in the absence of significant pulmonary disease, patients often have concurrent pulmonary disease, and at our center, we have identified a high prevalence (17 %) of ILD in the 29 HPS patients listed for LT with a MELD exception. These patients were able to obtain a MELD exception, because their severe hypoxemia was judged to be predominantly related to their HPS rather than their concurrent pulmonary disease. To our knowledge, this is the first case series focusing on ILD in patients with HPS undergoing LT. PFTs showed mostly mild restriction with a disproportionate decrease in the DLCO. HRCT showed lower lobe predominant subpleural reticulation, septal thickening, and mild architectural distortion without significant traction bronchiectasis or honeycombing, overall consistent with an NSIP pattern. However, the one postmortem histopathologic sample available from this series demonstrated subpleural fibrosis, honeycomb change, and fibroblastic foci, consistent with UIP. A few prior studies have noted concurrent ILD in patients with HPS. A retrospective study conducted by Gupta et al. from a tertiary care center found mild ILD in 4 (12 %) of 33 patients with HPS [5]. Another multicenter study of 218 patients undergoing evaluation for LT reported a higher incidence of ILD in patients with HPS: 7 versus 1 % in patients without HPS (p = 0.02) based on chest radiography [4]. The relationship between HPS and pulmonary fibrosis has yet to be elucidated. In our series, ILD was identified before initiation of oxygen, making the toxic effect of longterm oxygen therapy an unlikely etiology. Both cirrhosis and pulmonary fibrosis are fibrotic disorders characterized by epithelial/endothelial injury with dysregulated healing leading to fibroproliferation and tissue remodeling. Although the initial insult is different for these two organs, there may be considerable overlap in the pathogenesis of fibrosis [9]. The pathogenic hallmark of HPS is pulmonary precapillary and capillary dilation, which may be due to the decreased hepatic clearance of inflammatory and angiogenic cytokines. Animal models of HPS have demonstrated the involvement of several cytokines/growth factors, such as TNF-a, IL-1, TGF-b, endothelin-1, and VEGF [10, 11]. Importantly, these same cytokines/growth factors appear to be involved in the pathogenesis of ILDs [12–14]. There also may be factors among HPS patients that predispose them to developing lung injury. For example, HPS patients may have an increased incidence of chemical pneumonitis from
123
Lung (2014) 192:421–427
gastroesophageal reflux and aspiration due to increased intra-abdominal pressures or a tendency to sleep flat due to platypnea. Further observational and translational studies are required to better delineate the true link between these two clinical entities. Several studies have reported increased perioperative complications and mortality in patients with HPS, particularly in those with advanced HPS [7, 15–18]. One retrospective study found that 11 (57 %) of 23 patients with HPS had gas exchange impairment after LT [7]. Two of these patients had severe refractory hypoxemia and died from multiorgan failure. The remaining nine patients were weaned off oxygen over a median duration of 3 months [7]. Another study found that 5 (24 %) of 21 patients with HPS had hypoxemic respiratory failure that required prolonged mechanical ventilation posttransplant (median duration 18 days) [16]. Four (80 %) of these patients were weaned off oxygen at a median of 66 days. The study reported a 7 % 1-year mortality among patients with HPS and 14 % mortality among patients with severe HPS (PaO2 \ 50 mmHg) [16]. The outcomes in our HPS-ILD case series appear to be comparable to outcomes in these previously described HPS case series. All four patients with HPS and ILD who received LT at our center had perioperative complications. The three patients who received successful LT had postoperative hypoxemia that required prolonged mechanical ventilation. However, most were weaned off oxygen with no significant long-term sequelae. One patient developed an intraoperative air embolus, with resultant severe hypoxic respiratory failure and death. Our case series has a number of limitations given that it is a case series with a small sample size with all patients selected from one liver transplant center. Two of the patients identified as having ILD had no evidence of a significant restrictive pattern on PFT but were identified as having ILD based on evidence of clinically significant fibrosis on their HRCT. In addition, before 2008, patients underwent HRCT of the chest only if they had abnormalities on their chest radiograph, and it was not until 2008 that HRCT became part of the protocol on all patients with HPS. Thus, early pulmonary fibrosis could have been missed in some patients in the earlier era, resulting in an underestimation in the prevalence of ILD between 2000 and 2007. Lastly, this case series does not include those patients that were rejected from being placed on the LT list given the extent of their pulmonary fibrosis, creating an inherent selection bias in who is included in our case series. Despite these limitations, we did identify a high prevalence of ILD among patients with HPS who were listed for LT at our center with a MELD exception. The ILD in these patients has the radiographic appearance of NSIP, but our one histopathologic sample demonstrated features
Lung (2014) 192:421–427
consistent with UIP. There were perioperative complications with LT in all of these patients, but the surviving patients had eventual resolution of their hypoxemia with no evidence of ILD progression. The coexistence of HPS and ILD raises difficult questions regarding the appropriateness and timing of listing for liver transplantation. We believe that the presence of ILD should not be an absolute contraindication to LT in patients with HPS. However, when possible, patients with HPS-ILD should be monitored for ILD stability before LT. Conflict of interest
None.
References 1. Arguedas MR, Singh H, Faulk DK, Fallon MB (2007) Utility of pulse oximetry screening for hepatopulmonary syndrome. Clin Gastroenterol Hepatol 5(6):749–754 2. Lima BL, Franca AV, Pazin-Filho A et al (2004) Frequency, clinical characteristics, and respiratory parameters of hepatopulmonary syndrome. Mayo Clin Proc 79(1):42–48 3. Schenk P, Schoniger-Hekele M, Fuhrmann V, Madl C, Silberhumer G, Muller C (2003) Prognostic significance of the hepatopulmonary syndrome in patients with cirrhosis. Gastroenterology 125(4):1042–1052 4. Fallon MB, Krowka MJ, Brown RS et al (2008) Impact of hepatopulmonary syndrome on quality of life and survival in liver transplant candidates. Gastroenterology 135(4):1168–1175 5. Collisson EA, Nourmand H, Fraiman MH et al (2002) Retrospective analysis of the results of liver transplantation for adults with severe hepatopulmonary syndrome. Liver Transpl 8(10): 925–931 6. Deberaldini M, Arcanjo AB, Melo E et al (2008) Hepatopulmonary syndrome: morbidity and survival after liver transplantation. Transpl Proc 40(10):3512–3516
427 7. Taille C, Cadranel J, Bellocq A et al (2003) Liver transplantation for hepatopulmonary syndrome: a ten-year experience in Paris, France. Transplantation 75(9):1482–1489 discussion 1446-1487 8. Murray KF, Carithers RL Jr (2005) AASLD practice guidelines: evaluation of the patient for liver transplantation. Hepatology 41(6):1407–1432 9. Mercer PF, Chambers RC (2013) Coagulation and anticoagulation signaling in fibrosis. Biochim Biophys Acta 1832:1018–1027 10. Liu L, Liu N, Zhao Z et al (2012) TNF-a neutralization improves experimental hepatopulmonary syndrome in rats. Liver Int 32(6):1018–1026 11. Zhang J, Luo B, Tang L et al (2009) Pulmonary angiogenesis in a rat model of hepatopulmonary syndrome. Gastroenterology 136(3):1070–1080 12. Keane M, Strieter R, Lynch P, Belperio J (2006) Inflammation and angiogenesis in fibrotic lung disease. Sem Respir Crit Care Med 27(6):589–599 13. Kapanci Y, Desmouliere A, Pache JC, Redard M, Redard M, Gabbiani G (1995) Cytoskeletal protein modulation in pulmonary alveolar myofibroblasts during idiopathic pulmonary fibrosis. Possible role of transforming growth factor beta and tumor necrosis factor alpha. Am J Respir Crit Care Med 152:2163–2169 14. Michel RP, Stewart DJ, Sheppard M, Hamid Q, Corrin B (1993) Expression of endothelin-1 in lungs of patients with cryptogenic fibrosing alveolitis. Lancet 341(8860):1550–1554 15. Swanson KL, Wiesner RH, Krowka MJ (2005) Natural history of hepatopulmonary syndrome: Impact of liver transplantation. Hepatology 41(5):1122–1129 16. Gupta S, Castel H, Rao RV et al (2010) Improved survival after liver transplantation in patients with hepatopulmonary syndrome. Am J Transpl 10(2):354–363 17. Krowka MJ, Mandell MS, Ramsay MA et al (2004) Hepatopulmonary syndrome and portopulmonary hypertension: a report of the multicenter liver transplant database. Liver Transpl 10(2): 174–182 18. Arguedas MR, Abrams GA, Krowka MJ, Fallon MB (2003) Prospective evaluation of outcomes and predictors of mortality in patients with hepatopulmonary syndrome undergoing liver transplantation. Hepatology 37(1):192–197
123
