0
Clinical Investigations: TECHNIQUES |

Hepatopulmonary Syndrome*: A Prospective Study of Relationships Between Severity of Liver Disease, Pao2 Response to 100% Oxygen, and Brain Uptake After 99mTc MAA Lung Scanning FREE TO VIEW

Michael J. Krowka, MD, FCCP; Gregory A. Wiseman, MD; Omer L. Burnett, MD; James R. Spivey, MD; Terry Therneau, PhD; Michael K. Porayko, MD; Russell H. Wiesner, MD
Author and Funding Information

*From the Divisions of Pulmonary and Critical Care (Dr. Krowka) and Gastroenterology and Hepatology (Drs. Krowka, Parayko, and Wiesner), and the Departments of Diagnostic Radiology (Dr. Wiseman) and Health Sciences Research (Dr. Therneau), Mayo Clinic, Rochester, MN, and the Department of Diagnostic Radiology (Dr. Burnett) and the Division of Gastroenterology (Dr. Spivey), Mayo Clinic, Jacksonville, FL.

Correspondence to: Michael J. Krowka, MD, FCCP, Mayo Clinic, 200 1st St SW, Rochester, MN; e-mail: krowka@mayo.edu



Chest. 2000;118(3):615-624. doi:10.1378/chest.118.3.615
Text Size: A A A
Published online

Background: Because of the spectrum of intrapulmonary vascular dilation that characterizes hepatopulmonary syndrome (HPS), Pao2 while breathing 100% oxygen varies. Abnormal extrapulmonary uptake of 99mTc macroaggregated albumin (MAA) after lung perfusion is common.

Goal: To describe relationships between (1) severity of liver disease measured by the Child-Pugh (CP) classification; (2) Pao2 while breathing room air (RA) and 100% oxygen on 100% oxygen; and (3) extrapulmonary (brain) uptake of 99mTc MAA after lung scanning.

Methods and patients: We prospectively measured Pao2 on RA, Pao2 on 100% oxygen, and brain uptake after lung perfusion of 99mTc MAA in 25 consecutive HPS patients.

Results: Mean Pao2 on RA, Pao2 on 100% oxygen, Paco2 on RA, and 99mTc MAA brain uptake were similar when categorized by CP classification. Brain uptake was abnormal (≥ 6%) in 24 patients (96%). Brain uptake was 29 ± 20% (mean ± SD) and correlated inversely with Pao2 on RA (r = −0.57; p < 0.05) and Pao2 on 100% oxygen (r = −0.41; p < 0.05). Seven patients (28%) had additional nonvascular pulmonary abnormalities and lower Pao2 on 100% oxygen (215 ± 133 mm Hg vs 391 ± 137 mm Hg; p < 0.007). Eight patients (32%) died. Mortality in patients without coexistent pulmonary abnormalities was associated with greater brain uptake of 99mTc MAA (48 ± 18% vs 25 ± 20%; p < 0.04) and lower Pao2 on RA (40 ± 7 mm Hg vs 57 ± 11 mm Hg; p < 0.001).

Conclusion: The degree of hypoxemia associated with HPS was not related to the CP severity of liver disease. HPS patients with additional nonvascular pulmonary abnormalities exhibited lower Pao2 on 100% oxygen. Mortality was associated with lower Pao2 on RA, and with greater brain uptake of 99mTc MAA.

Figures in this Article

The term hepatopulmonary syndrome (HPS) refers to arterial hypoxemia caused by pulmonary vasodilation, which, in turn, is a consequence of either cirrhotic or noncirrhotic portal hypertension.12 Pathologic studies have demonstrated two types of pulmonary vasodilation: diffuse precapillary and capillary dilation34 and discrete arteriovenous communications.3 Most investigators accept the existence of these pulmonary vascular abnormalities if intrapulmonary vasodilation is suggested by either contrast-enhanced echocardiography58 or 99mTc macroaggregated albumin (MAA) lung scanning.,5,9

Such pulmonary vasodilation is associated with excess perfusion for a given ventilation, impaired diffusion-perfusion (incomplete penetration of oxygen through dilated vessels that abut alveoli), and true anatomic shunts (direct arteriovenous communications that bypass gas exchange units).34,1011 The resulting effect on arterial oxygenation ranges from an asymptomatic increase in the alveolar-arterial oxygen pressure difference (P[A-a]O2) to extreme breathlessness caused by severe hypoxemia as measured by Pao2 determined from standard arterial blood gas assessments. Abnormal oxygenation is frequently worse in the standing position (compared with supine), a phenomenon known as orthodeoxia. Position change in Pao2 caused by HPS may reflect the importance of cardiac output in maintaining Pao2 (reduced cardiac output in the standing position) as well as increasing lower lung vascular dilation when erect (more perfusion because of gravitational effects of standing).,1112 Importantly, patients with severe hypoxemia (Pao2 < 50 mm Hg) because of HPS have experienced increased mortality after liver transplantation; therefore, such patients represent a subgroup of particular interest.13

Our purpose in this study was to describe the clinical implications and relationships between Pao2 while breathing room air, Pao2 response to 100% inspired oxygen, extrapulmonary (brain) uptake of 99mTc MAA during lung perfusion scanning, and the severity of liver disease as measured by the Child-Pugh (CP) classification. We hypothesized that Pao2 while breathing 100% oxygen and brain uptake after lung perfusion would further characterize the severity of hypoxemia associated with HPS.

Written consent was provided by all patients participating in the clinical research protocols concerning HPS as approved by the Mayo Foundation Institutional Review Board. Patients were studied at Mayo institutions in Rochester, MN, and Jacksonville, FL, as outpatients in appropriate pulmonary function, cardiac echocardiography, and nuclear medicine departments.

Diagnostic Criteria for HPS

The diagnosis of HPS was established if each of three generally accepted hepatic, echocardiographic, and oxygenation criteria were met12: (1) chronic liver disease; (2) delayed positive-contrast echocardiography (left atrial microbubble opacification > 3 beats after right atrial opacification)7; and (3) abnormal oxygenation defined by Pao2 < 70 mm Hg,5,7,1417 or a P(A-a)O2 > 20 mm Hg assuming a respiratory quotient of 0.8.,17

Hepatic Criteria

Each patient had chronic liver disease characterized by clinical evidence of portal hypertension with or without cirrhosis. Portal hypertension was inferred if the patient had the following: (1) esophagogastric varices documented by esophagogastroduodenoscopy; (2) ascites by physical examination or ultrasonography; or (3) splenomegaly documented by CT scanning of the abdomen in the appropriate clinical setting with thrombocytopenia and leukopenia. Cirrhosis was established by liver biopsy or findings compatible with ultrasonography of the liver. Severity of liver disease was characterized by the CP scoring classification system: A ≤ 7 (least severe); B = 8 to 10; and C ≥ 11 (most severe). One to 3 points were assigned according to the increasing degree of abnormality in each of five variables (encephalopathy, ascites, serum total bilirubin, serum albumin, and prothrombin time or international normalized ratio at the time of lung perfusion scanning).18

Echocardiographic Criteria

Patients were determined to have intrapulmonary vascular dilations if results of transthoracic contrast-enhanced echocardiography were positive after the administration of 10 mL of hand-agitated normal saline solution in the supine position via an upper extremity peripheral vein. Positive was qualitatively defined as any visual opacification of the left heart chambers more than three cardiac cycles after appearance of microbubbles in the right ventricle.7 These findings suggested intrapulmonary passage of microbubbles through either dilated precapillary and capillary vessels or direct arteriovenous communications. No patient was found to have evidence of an intra-atrial right-to-left shunt (immediate opacification observed in the left atrium), that is, less than three cardiac cycles from the time of right atrial opacification.

Abnormal Oxygenation Criteria

Arterial blood gas samples were obtained from a single radial artery puncture while the patient was in a clinically stable situation. Pao2, Paco2, and pH were reported. The diagnosis of HPS was considered in patients with P(A-a)O2 > 20 mm Hg (abnormal oxygenation) or a Pao2 < 70 mm Hg breathing room air in any position at rest (supine, sitting, or standing). Those patients proceeded to the contrast-enhanced echocardiography evaluation and, if positive, had the formal standing oxygenation measurements and the lung scanning protocol.

Patient Selection and Prospective Study

From 1995 to mid-1999, 25 consecutive patients (Table 1 ) had a combination of abnormal arterial oxygenation and positive contrast-enhanced transthoracic echocardiography; therefore, each satisfied published criteria for HPS. These patients were identified through routine clinical evaluation conducted during liver transplant candidacy or hepatobiliary clinic consultation; Pao2 < 70 mm Hg resulted in subsequent echocardiographic examination. Prospectively, each patient gave written informed consent and subsequently underwent the expanded arterial blood gas studies (breathing 100% oxygen) and lung perfusion scanning as defined by an Institutional Review Board–approved protocol.

Pao2 Determinations While Breathing 100% Inspired Oxygen

A radial artery catheter was placed under local anesthetic, and each arterial sample (2 mL) was immediately analyzed (1,306 pH/blood gas analyzer; Instrumentation Laboratory; Lexington, MA). Pao2 measurements were obtained in the following sequence after assuming each position for a minimum of 10 min and maximum of 20 min: supine, breathing room air; standing, breathing room air; supine, breathing 100% oxygen; and standing, breathing 100% oxygen. All patients inspired 100% oxygen for 20 min supine. Breathing 100% oxygen was conducted by having the patient insert a sealed mouthpiece with a two-way demand value and wearing a snug nose clip. Pure (100%) oxygen was delivered via an E tank at an approximate rate of 25 L/min. Because arterial oxygen contents were not measured as part of this study, oxygen shunt estimates (or calculations based on assumed arteriovenous oxygen differences) were not studied.

99mTc MAA Lung and Brain Scanning

With the patient in the standing position for 10 min and breathing room air (within 24 h of the 100% inspired oxygen study), 2 mCi of 99mTc MAA (Dupont Pulmolite; Billerica, MA; 90% of the MAA particle size between 10 and 90 μm) was injected via a peripheral IV site. At 20 min after injection, quantitative brain imaging was conducted in the supine position, and a brain uptake percent (assuming a constant 13% blood flow to the brain) was obtained via the following calculation:

where GMT is the geometric mean counts of 99mTc MAA in the brain and lung. Images and GMT of the brain and lungs were obtained in the supine position, with each lateral head image obtained with a standard, large-field-of-view gamma camera fitted with a general purpose hole collimator; 20% window centered on the 140-keV photopeak of 99mTc for 5 min/view. Other than geometric mean calculations, no other corrections were made for attenuation.19 Abnormal uptake was defined as ≥ 6%.16

As a control group with advanced liver disease, but with pulmonary hemodynamics distinct from that seen in HPS, 12 patients with at least moderate portopulmonary hypertension (mean pulmonary artery pressure≥ 35 mm Hg and pulmonary vascular resistance ≥ 120 dyne · s · cm-5) underwent lung perfusion scanning via the same protocol.

Standard Pulmonary Function Tests

Pulmonary function tests (PFTs) were conducted according to American Thoracic Society standards for the measurement of lung volumes by plethysmography (total lung capacity and residual volume), expiratory airflows (ratio of FEV1 to FVC), and single breath diffusing capacity of the lung for carbon monoxide (Dlco) corrected for hemoglobin.

Pulmonary Angiography

At the discretion of the attending physician, bilateral pulmonary angiography was conducted to detect arteriovenous communications amenable to coil embolotherapy. Angiograms were classified as per our previous observations in patients with HPS.20

Type 1 angiograms were defined by either normal or diffuse spongy appearances in the arterial phase. Type 2 angiograms were defined as demonstrating distinct arteriovenous communications possibly amenable to coil embolotherapy.

Statistical Analysis

Results were reported as mean ± SD. Unpaired comparisons were analyzed using the t test; CP group comparisons were accomplished using one-way analysis of variance. Statistical significance was established at p < 0.05. Linear correlations were determined using the Pearson product-moment correlation coefficient.

A total of 25 patients (14 men/11 women; mean age, 50 years[ range, 12 to 68 years]) had HPS diagnosed. Individual data are shown in Table 1. Six patients had CP stage C liver disease, 9 were stage B, and 10 were stage A. Each patient had clinical findings consistent with portal hypertension. Seven of 25 patients (28%) had other clinically significant pulmonary abnormalities in addition to fulfilling criteria for the diagnosis of HPS (Table 2 ; patients 19 through 25).

PFTs were accomplished at our institution in 17 patients (68%); Dlco corrected for hemoglobin was abnormal (< 80% predicted) in 16 patients (94%). Mean Dlco was 57 ± 16% predicted (range, 33 to 98% predicted). Patients with total lung capacity < 80% predicted or FEV1/FVC < 70% are shown in Table 2. Six patients did not undergo PFTs at our institution because of the severity of their hepatic illness (patients 2 and 3), debilitating hepatic and respiratory illness (patient 25 had previous coronary artery bypass grafting, pulmonary fibrosis, and traction bronchiectasis and was not ambulatory), absence of respiratory symptoms (patients 14 and 15), or insurance limitations (patient 7); the last three patients had no respiratory symptoms. Two patients had PFTs accomplished elsewhere (patients 5 and 7), but data were not available for review. No patient had a current habit of smoking; 15 patients were previous smokers and 10 never smoked.

Pao2 While Breathing Room Air

While breathing room air, standing mean Pao2 was 50 ± 12 mm Hg (range, 33 to 74 mm Hg); 13 patients (52%) had severe hypoxemia (Pao2 < 50 mm Hg). Standing oxygenation was significantly was worse compared with supine oxygenation (Table 1). Mean Paco2 was similar in the supine and standing positions (32 ± 4 mm Hg; Table 1).

No significant differences in mean standing Pao2 were noted when patients were categorized by CP classification (Fig 1 , top). No difference was found when P(A-a)O2 was analyzed by CP classification (data not shown). Mean Pao2 and P(A-a)O2 (51 ± 12 mm Hg and 59 ± 17 mm Hg, respectively) in the 18 HPS patients without additional pulmonary abnormalities (Table 1; patients 1 through 18) were not significantly different in the standing position compared with the 7 patients with HPS and additional nonvascular pulmonary dysfunction (47 ± 12 mm Hg and 59 ± 15 mm Hg, respectively; p = 0.34).

Pao2 Response to 100% Inspired Oxygen

No significant difference was observed in Pao2 while breathing 100% oxygen in the standing position by CP classification (Fig 1, bottom). A significant, but weak, relationship between standing Pao2 while breathing room air and Pao2 while breathing 100% oxygen was noted (r = 0.52; p < 0.05; Fig 2 ).

The seven patients with additional pulmonary dysfunction had a mean standing Pao2 of 215 ± 133 mm Hg (range, 56 to 435 mm Hg) while breathing 100% oxygen, which was significantly less than the Pao2 while breathing 100% oxygen in the 18 HPS patients without additional lung problems (391 ±137 mm Hg; p < 0.007).

Pulmonary angiography was conducted in six patients with Pao2 > 300 mm Hg; no macroscopic lesions amenable to embolization were noted. Patients had either normal angiography or spongy vascular appearances (type 1 angiogram). Nine patients with Pao2 < 300 mm Hg underwent angiography, and two (patients 19 and 21) had embolization of discrete lesions (type 2 angiogram) with minimal improvement in oxygenation (change in Pao2 < 50 mm Hg while breathing 100% oxygen). The remaining seven patients had either diffuse, spongy arterial appearances or normal vasculature (type 1 angiograms).

99mTc MAA Lung Scanning and Brain Uptake

Each patient in the control group (patients with portopulmonary hypertension documented by right heart catheterization) had uptake< 6% (mean, 2.0 ± 1.0%; range, 0.4 to 3.9%). Twenty-four of 25 HPS patients (96%) had abnormal brain uptake of 99mTc MAA (Fig 3 ). Mean brain uptake was 29 ± 20% (range, 1 to 71%). No significant difference was noted in brain uptake percent categorized by the patient’s CP classification (Fig 4 ).

An inverse correlation between brain uptake and Pao2 while breathing 100% inspired oxygen was noted (r = −0.41; p < 0.05; Fig 5 , top). Relationships of similar strength were noted between Pao2 while breathing room air and brain uptake percent (r = −0.57; p < 0.05; Fig 5, bottom) and P(A-a)O2 and brain uptake percent (r = 0.59; p < 0.05). Correlation between Dlco and brain uptake percent was poor (r = 0.05). Although the mean brain uptake percent was slightly less in the seven HPS patients with additional pulmonary abnormalities (21 ± 14%) compared with the other 18 HPS patients (33 ± 22%), the difference did not reach statistical significance (p = 0.13).

Clinical and Mortality Correlates

There was no significant correlation between the Pao2 while breathing room air, P(A-a)O2, Pao2 while breathing 100% oxygen, or brain uptake of 99mTc MAA and other hepatic variables, such as serum total bilirubin, serum albumin, and platelet count (data not shown). Eight of 25 patients (32%) with HPS died during this study (Table 3 ). Four of 12 who underwent orthotopic liver transplantation (OLT) died within 3 months of the procedure; each had pre-OLT brain uptake> 30% (patients 5, 8, 11, and 12). Two patients (patients 22 and 24) had significant comorbid pulmonary conditions (severe pulmonary fibrosis; emphysema with bilateral hepatic hydrothorax), which precluded liver transplantation consideration. Two patients awaiting liver transplantation died of nonpulmonary complications of liver disease. In HPS patients without additional pulmonary abnormalities who died, nonsurvivors had significantly lower Pao2 (40 ± 7 mm Hg vs 57 ± 11 mm Hg; p < 0.0001), higher P(A-a)O2 (74 ± 10 mm Hg vs 51 ± 14 mm Hg; p < 0.0001), and greater brain uptake of 99mTc MAA (48 ± 18% vs 25 ± 20%; p < 0.04) compared with survivors.

Severity of Liver Disease

Each patient had advanced liver disease as demonstrated by clinical manifestations of portal hypertension, but we found no differences in oxygenation or brain uptake when severity of liver disease was further categorized by the CP classification. We expected the worst hypoxemia and greatest brain uptake in patients with CP class C severity liver disease, but that was not the case. Similarly, Abrams et al16 and Whyte et al21 noted that CP class C patients with HPS did not have the most severe hypoxemia or greatest brain uptake of 99mTc MAA. As a corollary, clinicians are reminded that severe HPS can occur in patients with mild liver disease (CP class A). Although each patient had clinical evidence of portal hypertension in this series, we should stress that our study did not address the severity of portal hypertension in relationship to oxygenation. Such investigation would further our liver-lung understanding and require invasive portal hemodynamic measurements.

Pao2 While Breathing 100% Oxygen

The limited correlation between Pao2 while breathing room air and Pao2 while breathing 100% oxygen suggested that additional information of clinical significance concerning interventional angiography and liver transplantation might be gathered from the Pao2 response to 100% inspired oxygen.

A favorable Pao2 while breathing 100% oxygen should virtually exclude the existence of a clinically significant anatomic (or physiologic) right to left shunt caused by discrete arteriovenous communications, in that very little improvement in Pao2 while breathing 100% oxygen would be expected in such situations. Deciding an appropriate Pao2 while breathing 100% oxygen cutoff to characterize as favorable is problematic; the limited data from this series showed that Pao2 > 300 mm Hg excluded the need for angiography. Therefore, the therapeutic benefit of conducting pulmonary angiography (for the purpose of arteriovenous communication embolization) would be minimal, and not advised in such patients. If coexistent pulmonary abnormalities cannot be found in patients with Pao2 while breathing 100% oxygen< 300 mm Hg, pulmonary angiography should be considered if therapeutic embolization is an option.2223

With reference to liver transplantation, one could hypothesize that patients with favorable Pao2 while breathing 100% oxygen might be expected to have the least complicated liver transplantation procedure from an anesthetic perspective and minimal respiratory-related mortality. Although a small series from Uemoto et al24 described less mortality in nine pediatric patients with HPS who underwent living-related OLT with Pao2 while breathing 100% oxygen> 300 mm Hg, only one of the four post-OLT deaths in our series had pre-OLT Pao2 while breathing 100% oxygen < 300 mm Hg.

Pao2 while breathing room air did not distinguish between HPS patients with and those without additional pulmonary pathologic processes. As recently reported by Martinez et al,25pulmonary abnormalities in addition to HPS may occur and contribute to hypoxemia. In our series, HPS patients with coexistent nonvascular pulmonary pathologic processes were not uncommon (28%). These patients had Pao2 responses to 100% inspired oxygen significantly less than those of HPS patients without additional pulmonary problems. The effects of compressive atelectasis caused by large pleural effusions (physiologic shunt with no ventilation to areas of perfusion),26 pulmonary fibrosis (additional oxygen diffusion limitation),27 and extensive secretions from bronchiectasis could result in mechanisms of hypoxemia that have limited response to 100% inspired oxygen. The combined effect of such pulmonary abnormalities, which are not simply ventilation-perfusion mismatches, with the physiology caused by HPS2527 may be responsible for our observations of Pao2 while breathing 100% oxygen. The effect of liver transplantation on arterial hypoxemia documented in the setting of additional nonvascular pulmonary abnormalities (such as pulmonary fibrosis or moderate COPD) is unknown.

Extrapulmonary (Brain) Uptake With 99mTc MAA Lung Scanning

The validity of quantifying right-to-left shunt through pulmonary vascular dilations (angiographically proven arteriovenous malformations) using 99mTc MAA lung perfusion scanning has been demonstrated by Chilvers et al.28 These authors reported a correlation of 0.993 in seven patients when comparing the degree of extrapulmonary uptake (over the right kidney) with shunt calculated by the 100% oxygen method. Abrams et al5,16 have reported the clinical utility of 99mTc MAA lung scanning in 25 patients with HPS (defined by abnormal contrast echocardiography). Extrapulmonary uptake over the brain was significantly increased (30 ± 4%) compared with control patients (group 1, 25 cirrhotic patients with normal contrast echocardiograms [2 ± 0.3%]; group 2, 15 patients with intrinsic lung disease, but without cirrhosis [2 ± 0.3%]).,16 Brain uptake correlations with Pao2 (r = −0.76) and P(A-a)O2 (r = 0.77) were slightly strongly than noted in our cohort. However, patient position while obtaining Pao2 or injecting the radioisotope was not reported. In our study, we correlated standing Pao2 obtained while breathing 100% oxygen and radioisotope injection while in the standing position in hopes of obtaining the most comparable and adverse oxygenation-extrapulmonary uptake relationship in the nonexercising state.

We suggest that these data support the notion that radioisotope lung scanning does not provide the physiologic information that may be inferred by 100% oxygen breathing in this syndrome. Indeed, the degree of extrapulmonary radioisotope uptake may reflect only the anatomic extent of pulmonary vascular dilation caused by increased precapillary and capillary diameters or anatomic shunts caused by direct arteriovenous communications. Pao2 measurements (breathing room air or 100% inspired oxygen) collectively quantify the total effect of all pulmonary pathologic processes upon oxygenation (vascular and nonvascular). Data from Whyte et al,21 in their study of eight HPS patients, would support this anatomic importance of 99mTc MAA lung scanning and its distinction from Pao2 determinations. Specifically, very large dilated channels or direct anatomic communications would result in concordance between brain uptake and measurements made breathing 100% oxygen. The channels are large enough to allow passage of 99mTc MAA and wide enough, even in the case of channels abutting alveoli, to impair the complete diffusion of oxygen from alveoli into the passing flow of venous blood. In the case of smaller dilations next to alveoli, passage of 99mTc MAA will still occur, but the diffusion of 100% oxygen into the capillary flow is not impaired and a discordance between the brain uptake (abnormal) and Pao2 while breathing 100% oxygen (normalization) occurs.

As suggested by Abrams et al,16 the degree of hypoxemia related to vascular dilation in patients with HPS (as opposed to coexistent nonvascular lung disorders such as expiratory airflow obstruction or pulmonary fibrosis) may best be quantified by extrapulmonary uptake estimates.

Concerning mortality after undergoing liver transplantation reported in patients with HPS who underwent pretransplant lung scanning, Uemoto et al24 reported that all three post-OLT deaths were associated with pre-OLT extrapulmonary uptake estimates of > 30%. In our series, each of the four patients studied before OLT who subsequently died after OLT had brain uptakes ≥ 30%. Additional data from Egawa et al29 reported on long-term survivals in 21 patients receiving transplants for biliary atresia with HPS; 1-year survival was 80%, 66%, and 50% associated with mild (< 20%), moderate (20 to 40%), and severe (> 40%) pre-OLT lung scanning shunt estimates, respectively. These data suggest the potential importance of 99mTc MAA for prognostic use; however, the procedure of 99mTc MAA scanning will need to be standardized for extrapulmonary uptake calculations.

Limitations

Limitations in this study include the subgroup of patients tested, the lack of correlation with other more invasive gas exchange techniques, and selection bias in the decision to proceed to contrast echocardiography or pulmonary angiography. Specifically, the techniques of 100% inspired oxygen described herein may not be applicable to the pediatric age group; except for one 12-year-old patient, our study included only adults. Referral bias in terms of patients with combined liver and severe lung problems must be considered. We recognize that invasive gas exchange study methods such as the multiple inert-gas elimination technique are perhaps most appropriate for mathematically analyzing factors contributing to hypoxemia. Even the multiple inert-gas elimination technique, however, conducted expertly by few centers,1112,3031 may pose difficult interpretation issues in cases of severe pulmonary vasodilation,31but comparisons to the 99mTc MAA method would be instructive. Such methods are not appropriate for practical clinical practice. In addition, we did not report shunt estimates using 100% inspired oxygen because those calculations would involve measuring arterial-venous content differences, which requires invasive sampling. Estimated content differences may be inaccurate in patients with advanced liver diseases because of extrapulmonary shunting.32 Finally, we followed published arterial blood gas criteria in selecting patients for contrast echocardiography. We may have inadvertently excluded subclinical HPS patients in this manner, but all patients with clinically significant hypoxemia associated with HPS were consecutively studied.

Additional quantification of the oxygenation abnormality in patients with HPS was obtained by (1) Pao2 response to 100% inspired oxygen and (2) extrapulmonary (brain) radioisotope uptake determinations. Severity of liver disease (measured by the CP classification) was not associated with either significantly different Pao2 determinations (breathing room air or 100% oxygen) or extrapulmonary uptake of 99mTc MAA. Pao2 response to 100% inspired oxygen was significantly worse in HPS patients with coexistent, nonvascular pulmonary pathologic processes. Mortality was associated with worse oxygenation breathing room air and greater extrapulmonary (brain) uptake of 99mTc MAA.

Abbreviations: CP = Child-Pugh; Dlco = single-breath diffusing capacity of the lung for carbon monoxide; HPS = hepatopulmonary syndrome; MAA = macroaggregated albumin; OLT = orthotopic liver transplantation; P(A-a)O2 = alveolar-arterial oxygen pressure difference; PFT = pulmonary function test

Table Graphic Jump Location
Table 1. Hepatic, Oxygenation, and 99mTc MAA Brain Uptake Data in Patients With HPS
* 

Hepc = hepatitis C; Bat = bilary atresia; Hem = hemachromatosis; Etoh = alcoholic cirrhosis; Wils = Wilson’s disease; Nph = noncirrhotic portal hypertension; Nash = nonalcoholic steatohepatitis; Psc = primary sclerosing cholangitis; Iduc = idiopathic ductopenia; Cryp = cryptogenic cirrhosis; Dx = diagnosis; Ang = pulmonary angiography; Tx = liver transplantation; y = yes, procedure accomplished; n = no, procedure not done; RA = room air.

 

p < 0.05 vs supine mean values.

Table Graphic Jump Location
Table 2. Coexistent Pulmonary Dysfunction in Patients With HPS*
* 

TLC = total lung capacity by plethysmography; FEV1% = ratio of FEV1/FVC; RV = residual volume.

 

Values corrected for serum hemoglobin.

Figure Jump LinkFigure 1. Oxygenation categorized by the CP classification for the severity of liver disease. Top: Pao2 while breathing room air, standing position. Bottom: Pao2 while breathing 100% oxygen, standing position. No significant differences were noted when mean values were compared among CP groups.Grahic Jump Location
Figure Jump LinkFigure 2. Relationship between Pao2 while breathing room air with Pao2 while breathing 100% oxygen (both measurements in the standing position). Note that when Pao2 was < 50 mm Hg, there was a significant variability in Pao2 response to 100% oxygen.Grahic Jump Location
Figure Jump LinkFigure 3. Extrapulmonary radionuclide (99mTc MAA) uptake demonstrated over the brain (patient 10). Ant = anterior; Post = posterior; Left Lat = left lateral; Right Lat = right lateral.Grahic Jump Location
Figure Jump LinkFigure 4. Extrapulmonary (brain) uptake of 99mTc MAA (injected in the standing position) categorized by the CP classification for severity of liver disease; no significant differences noted among CP groups.Grahic Jump Location
Figure Jump LinkFigure 5. Relationship between extrapulmonary 99mTc MAA uptake percent (injected in the standing position) over the brain and Pao2. Top: Pao2 breathing 100% oxygen, standing position. Bottom: Pao2 breathing room, standing position.Grahic Jump Location
Table Graphic Jump Location
Table 3. Mortality Summary
Lange, PA, Stoller, JK (1995) The hepatopulmonary syndrome.Ann Intern Med122,521-529. [PubMed]
 
Hervé, P, Lebrec, D, Brenot, F, et al Pulmonary vascular disorders in portal hypertension.Eur Respir J1998;11,1153-1166. [CrossRef] [PubMed]
 
Rydell, R, Hoffbauer, FW Multiple pulmonary arteriovenous fistulae in juvenile cirrhosis.Am J Med1956;21,450-459. [CrossRef] [PubMed]
 
Berthelot, P, Walker, JG, Sherlock, S, et al Arterial changes in the lungs in cirrhosis of the liver-lung spider nevi.N Engl J Med1966;274,291-298. [CrossRef] [PubMed]
 
Abrams, GA, Jaffe, CC, Hoffer, PB, et al Diagnostic utility of contrast echocardiography and lung perfusion scan in patients with hepatopulmonary syndrome.Gastroenterology1995;109,1283-1288. [CrossRef] [PubMed]
 
Hind, CRK, Wong, CM Detection of pulmonary arteriovenous fistulae in patients with cirrhosis by contrast enhanced echocardiography.Gut1981;22,1042-1045. [CrossRef] [PubMed]
 
Krowka, MJ, Tajik, AJ, Dickson, ER, et al Intrapulmonary vascular dilatations in liver transplant candidates: screening by two-dimensional contrast enhanced echocardiography.Chest1990;97,1165-1170. [CrossRef] [PubMed]
 
Hopkins, WE, Waggnoner, AD, Barzilai, B Frequency and significance of intrapulmonary right to left shunting in end stage hepatic disease.Am J Cardiol1992;70,516-519. [CrossRef] [PubMed]
 
Wolfe, JD, Tashkin, DP, Holly, FE, et al Hypoxemia of cirrhosis: detection of abnormal small pulmonary vascular channels by a quantitative radionuclide method.Am J Med1977;63,746-754. [CrossRef] [PubMed]
 
Thorens, JB, Junod, AF Hypoxemia and liver cirrhosis: a new argument in favor of a “diffusion-perfusion defect.”Eur Respir J1992;5,754-756. [PubMed]
 
Edell, ES, Cortese, DA, Krowka, MJ, et al Severe hypoxemia and liver disease.Am Rev Respir Dis1989;140,1631-1635. [CrossRef] [PubMed]
 
Andrivet, P, Cadranel, J, Housset, B, et al Mechanisms of impaired oxygenation in patients with liver cirrhosis and severe respiratory insufficiency: effects of indomethacin.Chest1993;103,500-507. [CrossRef] [PubMed]
 
Krowka, MJ, Porayko, MK, Plevak, DJ, et al Hepatopulmonary syndrome with progressive hypoxemia as an indication for liver transplantation: case reports and literature review.Mayo Clin Proc1997;72,44-53. [CrossRef] [PubMed]
 
Moller, S, Hillings, J, Christensen, E, et al Arterial hypoxaemia in cirrhosis: fact or fiction?Gut1998;42,868-874. [CrossRef] [PubMed]
 
Vachiéry, F, Moreau, R, Hadengue, A, et al Hypoxemia in patients with cirrhosis: relationship with liver failure and hemodynamic alterations.J Hepatol1997;27,492-495. [CrossRef] [PubMed]
 
Abrams, GA, Nanda, NC, Dubovsky, EV, et al Use of macroaggregated lung perfusion scan to diagnose hepatopulmonary syndrome: a new approach.Gastroenterology1998;114,305-310. [CrossRef] [PubMed]
 
Baum GL, Wolinsky E. In: Textbook of pulmonary diseases. 5th ed. Boston, MA: Little Brown & Co Inc, 1994; 155.
 
Henderson, JM Surgical management of portal hypertension.Schiff ER, Sorrell MF, Maddrey WC1998,445 Lippincott-Raven. Schiff’s diseases of the liver. 8th ed. Philadelphia, PA:
 
Barnes, EW In vivoquantification of activity by planar imaging. Henkin, RE eds.Nuclear medicine1996,216-218 Mosby. St. Louis, MO:
 
Krowka, MJ, Dickson, ER, Cortese, DA Hepatopulmonary syndrome: clinical observations and lack of therapeutic response to somatostatin analogue.Chest1993;104,515-521. [CrossRef] [PubMed]
 
Whyte, MKB, Hughes, JMB, Peters, AM, et al Analysis of intrapulmonary right to left shunt in hepatopulmonary syndrome.J Hepatol1998;29,85-93. [CrossRef] [PubMed]
 
McAdams, HP, Erasmus, J, Crockett, R, et al The hepatopulmonary syndrome: radiologic findings with 10 patients.AJR Am J Roentgenol1996;166,1379-1385. [PubMed]
 
Poterucha, JJ, Krowka, MJ, Dickson, ER, et al Failure of hepatopulmonary syndrome to resolve after liver transplantation and successful treatment with embolotherapy.Hepatology1995;21,96-100. [CrossRef] [PubMed]
 
Uemoto, S, Inomata, Y, Egawa, H, et al Effects of hypoxemia on early postoperative course of liver transplantation in pediatric patients with intrapulmonary shunting.Transplantation1997;63,407-414. [CrossRef] [PubMed]
 
Martinez, G, Barbera, JA, Navasa, M, et al Hepatopulmonary syndrome associated with cardiorespiratory disease.J Hepatol1999;30,882-889. [CrossRef] [PubMed]
 
Agustí, AGN, Cardus, J, Roca, J, et al Ventilation-perfusion mismatch in patients with pleural effusion.Am J Respir Crit Care Med1997;156,1205-1209. [PubMed]
 
Agustí, AGN, Roca, J, Gea, J, et al Mechanisms of gas exchange impairment in idiopathic pulmonary fibrosis.Am Rev Respir Dis1991;143,219-225. [PubMed]
 
Chilvers, ER, Peters, AM, George, P, et al Quantification of right to left shunt through pulmonary arteriovenous malformations using99Tcmalbumin microspheres.Clin Radiol1988;39,611-614. [CrossRef] [PubMed]
 
Egawa, H, Kasahara, M, Inomata, Y, et al Long-term outcome of living related liver transplantation for patients with intrapulmonary shunting and strategy for complications.Transplantation1999;67,712-717. [CrossRef] [PubMed]
 
Rodriguez-Roisin, R, Roca, J, Agusti, AGN, et al Gas exchange and pulmonary vascular reactivity in patients with liver cirrhosis.Am Rev Respir Dis1987;135,1085-1092. [PubMed]
 
Crawford, ABH, Regnis, J, Laks, L, et al Pulmonary vascular dilatation and diffusion-dependent impairment of gas exchange in liver cirrhosis.Eur Respir J1995;8,2015-2021. [CrossRef] [PubMed]
 
Fernandez-Rodriguez, CM, Prieto, J, Zozaya, JM, et al Arteriovenous shunting, hemodynamic changes, and renal sodium retention in liver cirrhosis.Gastroenterology1993;104,1139-1145. [PubMed]
 

Figures

Figure Jump LinkFigure 1. Oxygenation categorized by the CP classification for the severity of liver disease. Top: Pao2 while breathing room air, standing position. Bottom: Pao2 while breathing 100% oxygen, standing position. No significant differences were noted when mean values were compared among CP groups.Grahic Jump Location
Figure Jump LinkFigure 2. Relationship between Pao2 while breathing room air with Pao2 while breathing 100% oxygen (both measurements in the standing position). Note that when Pao2 was < 50 mm Hg, there was a significant variability in Pao2 response to 100% oxygen.Grahic Jump Location
Figure Jump LinkFigure 3. Extrapulmonary radionuclide (99mTc MAA) uptake demonstrated over the brain (patient 10). Ant = anterior; Post = posterior; Left Lat = left lateral; Right Lat = right lateral.Grahic Jump Location
Figure Jump LinkFigure 4. Extrapulmonary (brain) uptake of 99mTc MAA (injected in the standing position) categorized by the CP classification for severity of liver disease; no significant differences noted among CP groups.Grahic Jump Location
Figure Jump LinkFigure 5. Relationship between extrapulmonary 99mTc MAA uptake percent (injected in the standing position) over the brain and Pao2. Top: Pao2 breathing 100% oxygen, standing position. Bottom: Pao2 breathing room, standing position.Grahic Jump Location

Tables

Table Graphic Jump Location
Table 1. Hepatic, Oxygenation, and 99mTc MAA Brain Uptake Data in Patients With HPS
* 

Hepc = hepatitis C; Bat = bilary atresia; Hem = hemachromatosis; Etoh = alcoholic cirrhosis; Wils = Wilson’s disease; Nph = noncirrhotic portal hypertension; Nash = nonalcoholic steatohepatitis; Psc = primary sclerosing cholangitis; Iduc = idiopathic ductopenia; Cryp = cryptogenic cirrhosis; Dx = diagnosis; Ang = pulmonary angiography; Tx = liver transplantation; y = yes, procedure accomplished; n = no, procedure not done; RA = room air.

 

p < 0.05 vs supine mean values.

Table Graphic Jump Location
Table 2. Coexistent Pulmonary Dysfunction in Patients With HPS*
* 

TLC = total lung capacity by plethysmography; FEV1% = ratio of FEV1/FVC; RV = residual volume.

 

Values corrected for serum hemoglobin.

Table Graphic Jump Location
Table 3. Mortality Summary

References

Lange, PA, Stoller, JK (1995) The hepatopulmonary syndrome.Ann Intern Med122,521-529. [PubMed]
 
Hervé, P, Lebrec, D, Brenot, F, et al Pulmonary vascular disorders in portal hypertension.Eur Respir J1998;11,1153-1166. [CrossRef] [PubMed]
 
Rydell, R, Hoffbauer, FW Multiple pulmonary arteriovenous fistulae in juvenile cirrhosis.Am J Med1956;21,450-459. [CrossRef] [PubMed]
 
Berthelot, P, Walker, JG, Sherlock, S, et al Arterial changes in the lungs in cirrhosis of the liver-lung spider nevi.N Engl J Med1966;274,291-298. [CrossRef] [PubMed]
 
Abrams, GA, Jaffe, CC, Hoffer, PB, et al Diagnostic utility of contrast echocardiography and lung perfusion scan in patients with hepatopulmonary syndrome.Gastroenterology1995;109,1283-1288. [CrossRef] [PubMed]
 
Hind, CRK, Wong, CM Detection of pulmonary arteriovenous fistulae in patients with cirrhosis by contrast enhanced echocardiography.Gut1981;22,1042-1045. [CrossRef] [PubMed]
 
Krowka, MJ, Tajik, AJ, Dickson, ER, et al Intrapulmonary vascular dilatations in liver transplant candidates: screening by two-dimensional contrast enhanced echocardiography.Chest1990;97,1165-1170. [CrossRef] [PubMed]
 
Hopkins, WE, Waggnoner, AD, Barzilai, B Frequency and significance of intrapulmonary right to left shunting in end stage hepatic disease.Am J Cardiol1992;70,516-519. [CrossRef] [PubMed]
 
Wolfe, JD, Tashkin, DP, Holly, FE, et al Hypoxemia of cirrhosis: detection of abnormal small pulmonary vascular channels by a quantitative radionuclide method.Am J Med1977;63,746-754. [CrossRef] [PubMed]
 
Thorens, JB, Junod, AF Hypoxemia and liver cirrhosis: a new argument in favor of a “diffusion-perfusion defect.”Eur Respir J1992;5,754-756. [PubMed]
 
Edell, ES, Cortese, DA, Krowka, MJ, et al Severe hypoxemia and liver disease.Am Rev Respir Dis1989;140,1631-1635. [CrossRef] [PubMed]
 
Andrivet, P, Cadranel, J, Housset, B, et al Mechanisms of impaired oxygenation in patients with liver cirrhosis and severe respiratory insufficiency: effects of indomethacin.Chest1993;103,500-507. [CrossRef] [PubMed]
 
Krowka, MJ, Porayko, MK, Plevak, DJ, et al Hepatopulmonary syndrome with progressive hypoxemia as an indication for liver transplantation: case reports and literature review.Mayo Clin Proc1997;72,44-53. [CrossRef] [PubMed]
 
Moller, S, Hillings, J, Christensen, E, et al Arterial hypoxaemia in cirrhosis: fact or fiction?Gut1998;42,868-874. [CrossRef] [PubMed]
 
Vachiéry, F, Moreau, R, Hadengue, A, et al Hypoxemia in patients with cirrhosis: relationship with liver failure and hemodynamic alterations.J Hepatol1997;27,492-495. [CrossRef] [PubMed]
 
Abrams, GA, Nanda, NC, Dubovsky, EV, et al Use of macroaggregated lung perfusion scan to diagnose hepatopulmonary syndrome: a new approach.Gastroenterology1998;114,305-310. [CrossRef] [PubMed]
 
Baum GL, Wolinsky E. In: Textbook of pulmonary diseases. 5th ed. Boston, MA: Little Brown & Co Inc, 1994; 155.
 
Henderson, JM Surgical management of portal hypertension.Schiff ER, Sorrell MF, Maddrey WC1998,445 Lippincott-Raven. Schiff’s diseases of the liver. 8th ed. Philadelphia, PA:
 
Barnes, EW In vivoquantification of activity by planar imaging. Henkin, RE eds.Nuclear medicine1996,216-218 Mosby. St. Louis, MO:
 
Krowka, MJ, Dickson, ER, Cortese, DA Hepatopulmonary syndrome: clinical observations and lack of therapeutic response to somatostatin analogue.Chest1993;104,515-521. [CrossRef] [PubMed]
 
Whyte, MKB, Hughes, JMB, Peters, AM, et al Analysis of intrapulmonary right to left shunt in hepatopulmonary syndrome.J Hepatol1998;29,85-93. [CrossRef] [PubMed]
 
McAdams, HP, Erasmus, J, Crockett, R, et al The hepatopulmonary syndrome: radiologic findings with 10 patients.AJR Am J Roentgenol1996;166,1379-1385. [PubMed]
 
Poterucha, JJ, Krowka, MJ, Dickson, ER, et al Failure of hepatopulmonary syndrome to resolve after liver transplantation and successful treatment with embolotherapy.Hepatology1995;21,96-100. [CrossRef] [PubMed]
 
Uemoto, S, Inomata, Y, Egawa, H, et al Effects of hypoxemia on early postoperative course of liver transplantation in pediatric patients with intrapulmonary shunting.Transplantation1997;63,407-414. [CrossRef] [PubMed]
 
Martinez, G, Barbera, JA, Navasa, M, et al Hepatopulmonary syndrome associated with cardiorespiratory disease.J Hepatol1999;30,882-889. [CrossRef] [PubMed]
 
Agustí, AGN, Cardus, J, Roca, J, et al Ventilation-perfusion mismatch in patients with pleural effusion.Am J Respir Crit Care Med1997;156,1205-1209. [PubMed]
 
Agustí, AGN, Roca, J, Gea, J, et al Mechanisms of gas exchange impairment in idiopathic pulmonary fibrosis.Am Rev Respir Dis1991;143,219-225. [PubMed]
 
Chilvers, ER, Peters, AM, George, P, et al Quantification of right to left shunt through pulmonary arteriovenous malformations using99Tcmalbumin microspheres.Clin Radiol1988;39,611-614. [CrossRef] [PubMed]
 
Egawa, H, Kasahara, M, Inomata, Y, et al Long-term outcome of living related liver transplantation for patients with intrapulmonary shunting and strategy for complications.Transplantation1999;67,712-717. [CrossRef] [PubMed]
 
Rodriguez-Roisin, R, Roca, J, Agusti, AGN, et al Gas exchange and pulmonary vascular reactivity in patients with liver cirrhosis.Am Rev Respir Dis1987;135,1085-1092. [PubMed]
 
Crawford, ABH, Regnis, J, Laks, L, et al Pulmonary vascular dilatation and diffusion-dependent impairment of gas exchange in liver cirrhosis.Eur Respir J1995;8,2015-2021. [CrossRef] [PubMed]
 
Fernandez-Rodriguez, CM, Prieto, J, Zozaya, JM, et al Arteriovenous shunting, hemodynamic changes, and renal sodium retention in liver cirrhosis.Gastroenterology1993;104,1139-1145. [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
PubMed Articles
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543