Interact CardioVasc Thorac Surg 2007;6:720-730. doi:10.1510/icvts.2007.162487
© 2007 European Association of Cardio-Thoracic Surgery
Institutional report - Pulmonary |
Factors predicting early postoperative liver cirrhosis-related complications after lung cancer surgery in patients with liver cirrhosis ,
Takashi Iwata*,
Kiyotoshi Inoue,
Noritoshi Nishiyama,
Koshi Nagano,
Nobuhiro Izumi,
Takuma Tsukioka,
Shoji Hanada and
Shigefumi Suehiro
Department of Thoracic Surgery, Osaka City University Graduate School of Medicine, 1-4-3 Asahi-machi, Abeno-ku, Osaka 545-8585, Japan
Received 28 June 2007;
received in revised form 7 August 2007;
accepted 10 August 2007
 This study was funded by Osaka City University Graduate School of Medicine. 
Financial or other potential conflicts of interest do not exist.
*Corresponding author. Tel.: +81-6-6645-3841; fax: +81-6-6646-6057.
E-mail address: taiwata{at}med.osaka-cu.ac.jp (T. Iwata).
 |
Abstract
|
|---|
We aimed to determine the factors predicting liver cirrhosis-related complications in the early postoperative period after lung cancer surgery in patients with liver cirrhosis. We retrospectively reviewed the medical records of patients who underwent curative surgery for primary lung cancer in our institute from January 1990 to March 2007, finding 37 cases with comorbid liver cirrhosis. These patients were divided into two groups, according to whether liver failure, bleeding, and critical infection had occurred postoperatively. Various clinical parameters were analyzed statistically between the bigeminal groups. Liver cirrhosis-related complications occurred in seven of the 37 patients (18.9%). Transient liver failure occurred in two patients (5.4%) after pulmonary resection. Acute intrathoracic bleeding occurred in four cases (10.8%). Two patients died (5.4%) in both cases due to sepsis. Preoperative total bilirubin (P<0.05), and indocyanine green retention rate at 15 min (P<0.05) were significantly higher in patients with liver failure. Only serum value of total bilirubin was an independent risk factor (P<0.05) by multivariate analysis. In predicting death from infection, only preoperative nutritional status was a significant risk factor (P<0.05). To avoid postoperative cirrhosis-related complications, preoperative preparation to improve their liver function and nutrition status is essential.
Key Words: Liver cirrhosis; Liver failure; Lung cancer; Pulmonary resection
|
1. Introduction
|
|---|
Liver cirrhosis is a highly lethal condition by itself. Its life expectancy is <40% of that of the same-aged general population [1]. Furthermore, liver cirrhosis often complicates hepatocellular carcinoma (HCC) during its clinical course and this may also shorten the life expectancy. However, when proper treatment has been applied, the 5-year survival rate is >50% [2]. On the other hand, lung cancer is considered more lethal and has shorter life expectancy than liver cirrhosis even with HCC. Therefore, for patients with both lung cancer and liver cirrhosis, treatment of lung cancer would have an affect on overall survival and may benefit the patients, if peri-operative cirrhosis-related complications are avoidable. However, safer methodology of pulmonary resection to overcome these complications in such cirrhotic patients is unknown.
Patients with liver cirrhosis have various metabolic dysfunctions that can be problematic when considering surgical intervention for the following three reasons: (1) poor tolerance of invasive stress such as surgical procedures; (2) diminished coagulability [3]; and (3) increased susceptibility to infection [4–6]. These problems may lead to three major complications following surgery in cirrhotic patients. (1) In patients with cirrhosis, surgical stress beyond the impaired liver function may provoke liver failure postoperatively, with hepatic encephalopathy, ascites, and jaundice. (2) Acute bleeding may occur from the operative field and/or suddenly from the upper gastrointestinal tract due to rupture of esophagogastric varices or gastroduodenal ulcers. (3) Postoperative acute severe infection may occur due to the altered immunological status as a consequence of malnutrition and metabolic abnormalities. In this study, we aimed to determine the risk factors for cirrhosis-related complications such as liver failure, acute bleeding, and severe infections as referenced earlier, in the early postoperative period after lung cancer surgery in patients with liver cirrhosis.
|
2. Materials and methods
|
|---|
2.1. Patients
We retrospectively reviewed the medical records of 928 patients who underwent curative surgical treatment for primary lung cancer from January 1990 to March 2007 in Osaka City University Medical School Hospital. Thirty-seven cases of lung cancer surgery in patients with liver cirrhosis were found. Informed consent for the use of their clinical data in later clinical study was obtained at time of pulmonary resection from each individual patient and the institutional review board approved this retrospective study. Liver cirrhosis had been diagnosed in each patient according to physical findings, laboratory tests, and morphological findings on ultrasonography and computed tomography by two or more hepatologists preoperatively. Pulmonary resection had been performed when liver function was maintained as Child–Pugh class A or B, even with viable and/or multiple HCC. Patients who exhibited jaundice, increased ascites, flapping tremor, disorientation, or decreased mental function after pulmonary resection were included in the liver failure group. Patients without any of these symptoms postoperatively were regarded as the non-liver failure group. Patients who needed a blood transfusion after pulmonary resection due to massive bleeding from the chest tube and/or upper gastrointestinal tract were included in the bleeding group. Patients who did not need a blood transfusion postoperatively were regarded as the non-bleeding group. Patients who died due to complicated severe infection after pulmonary resection were included in the infection death group. Patients without such postoperative severe infection were regarded as the non-infection group.
2.2. Measurements
Clinical information about the patients' profile, results of laboratory and pulmonary function tests, lung disease, performed operative method, and the outcome were investigated in detail.
2.3. Statistics
Data were compared between groups by the following methods. Student's t-test was used to compare patient ages. Pearson's  2-test was used to compare categorical data between groups. Yates' continuity correction was calculated when the expected value was five or less. Fisher's exact test was used to compare other categorical data. The Mann–Whitney U-test was used to analyze differences in laboratory data.
|
3. Results
|
|---|
3.1. Cause of liver cirrhosis
As a cause of liver cirrhosis, hepatitis C virus (HCV) infection was most common and seen in 30 cases (81.1%); 27 of them were infected with HCV alone (73.0%) and the remaining 3 (8.1%) was mixed with comorbid hepatitis B virus (HBV) infection. Infection with HBV alone was found in only one case (2.7%) and the remaining six were without evidence of HCV and HBV infections. Of these six cases, five (13.5%) were found to be pure alcoholic cirrhosis and one (2.7%) was non-B non-C cirrhosis. Habitual alcohol consumption was seen in 22 cases.
3.2. Incidence and outcome of liver cirrhosis-related complications after lung cancer surgery
Details of the patients who developed postoperative complications due to liver cirrhosis are shown in Table 1. Of the 37 cases that underwent curative surgery for lung cancer, liver cirrhosis-related complications occurred in seven (18.9%). Two patients died (5.4%), one was a postoperative death (death within 30 days after pulmonary resection) (2.7%) and one in-hospital death (death more than 30 days after pulmonary resection) (2.7%). Both cases were due to infectious disease. Patient 1 died of sepsis and following acute respiratory failure on the sixth postoperative day. This patient had untreated severe diabetes and her preoperative nutritional status was not good. Patient 2 also died of sepsis and following multiple organ failure on day 46. On day 5, sudden shock presented with high fever. The systemic status worsened rapidly, though hemodialysis had been started for acute renal failure. These two cases were included in the infection death group. The other 35 were treated as the non-infection group.
Liver failure occurred in two of the 37 patients (5.4%) in the early period after pulmonary resection. Patients 3 and 4 exhibited slightly reduced consciousness and a negligible flapping tremor on postoperative day 6 and day 7, respectively. Increased ascites and serum total bilirubin (T-BIL) were observed and hepatic coma due to liver failure was diagnosed on day 8 and day 11, respectively. Both of these patients had a high indocyanine green retention rate at 15 min (ICGR15), preoperatively, and recovered well after liver-supporting therapy and were discharged without any clinical and physical findings of liver failure. These two cases were included in the liver failure group and the remaining 35 were regarded as the non-liver failure group.
Acute postoperative intrathoracic bleeding occurred in four of the 37 patients (10.8%). Patient 5 needed re-thoracotomy to control the bleeding. Patients 4, 6, and 7 needed blood transfusion including fresh frozen plasma and/or platelets (8.1%). These four patients had exhibited reduced platelet counts preoperatively. Patient 7 also had simultaneous bleeding from a gastroduodenal ulcer (2.7%). In all four cases, the bleeding was controlled successfully after therapy and the patients were discharged without any further events. These four cases were included in the bleeding group and the remaining 33 were regarded as the non-bleeding group.
3.3. Clinical profiles of patients and risk factors predicting liver failure in the early postoperative period after lung cancer surgery
Clinical profiles and the factors predicting postoperative liver failure are shown in Table 2. ICGR15 was measured in 25 cases. Preoperative serum alpha-fetoprotein (AFP) (P<0.01), T-BIL (P<0.05), and ICGR15 (P<0.05) were significantly higher in the liver failure group. The prevalence of HCC, higher serum alanine aminotransferase, decreased values on the prothrombin test, and reduced platelet count was comparatively greater in the liver failure group.
View this table:
[in this window]
[in a new window]
|
Table 2 Clinical profiles and factors predicting postoperative liver failure after lung cancer surgery in patients associated with liver cirrhosis
|
|
The results of univariate analysis of factors influencing early postoperative liver failure are shown in Table 3. No significant factor was found in the measured parameters. Serum AFP and T-BIL were comparatively higher in the liver failure group than in the non-liver failure group. Table 4 shows the results of multivariate analysis to evaluate how limited surgery, mediastinal dissection, Child–Pugh class, and preoperative serum AFP and T-BIL affected early postoperative liver failure. Only T-BIL was an independent risk factor (P<0.05). The odds ratio was about 1.6 per 0.1 mg/dl of the T-BIL value. AFP also tended to predict liver failure.
View this table:
[in this window]
[in a new window]
|
Table 3 Result of univariate analysis for factors predicting postoperative liver failure after lung cancer surgery in patients associated with liver cirrhosis
|
|
View this table:
[in this window]
[in a new window]
|
Table 4 Liver damage score and serum alpha-fetoprotein value in predicting postoperative liver failure after lung cancer surgery in patients associated with liver cirrhosis adjusting lung cancer staging and operative invasiveness calculated by mulivariate analysis
|
|
3.4. Factors predicting postoperative bleeding tendency after lung cancer surgery in patients with liver cirrhosis
Clinical profiles and the factors predicting postoperative bleeding are shown in Table 5. The T factor of the lung disease (P<0.05) and the size of the tumor (P<0.05) were correlated with postoperative bleeding. Platelet count and arterial partial pressure of oxygen were comparatively lower in the bleeding group. By univariate analysis, as shown in Table 6, wedge resection for the lung cancer was a significant risk factor (P<0.05) and its odds ratio was ten times higher than that for lobectomy. The T factor was a negative significant risk factor (P<0.05). The odds ratio of T1 disease against T2 or more was 0.046.
View this table:
[in this window]
[in a new window]
|
Table 5 Clinical profiles and factors predicting postoperative bleeding after lung cancer surgery in patients associated with liver cirrhosis
|
|
View this table:
[in this window]
[in a new window]
|
Table 6 Result of univariate analysis for factors predicting postoperative bleeding after lung cancer surgery in patients associated with liver cirrhosis
|
|
3.5. Factors predicting postoperative lethal infection after lung cancer surgery in patients with liver cirrhosis
Table 7 shows the patients' clinical profiles and the results of calculations performed to predict critical postoperative infectious disease. No significant factors were found. According to the univariate analysis shown in Table 8, only preoperative nutritional status was a statistically significant risk factor predicting postoperative death from infection (P<0.05). The odds ratio was 34.
View this table:
[in this window]
[in a new window]
|
Table 7 Clinical profiles and factors predicting postoperative Infectious death after lung cancer surgery in patients associated with liver cirrhosis
|
|
View this table:
[in this window]
[in a new window]
|
Table 8 Result of univariate analysis for factors predicting postoperative bleeding after lung cancer surgery in patients associated with liver cirrhosis
|
|
|
4. Discussion
|
|---|
Comparison of invasiveness of different surgical procedures on different organs is difficult, thoracic surgery may be less invasive to the cirrhotic liver than open abdominal surgery. Shimada and coworkers reported that postoperative complications were less common following transdiaphragmatic hepatic surgery for HCC in cirrhotic patients than following conventional transabdominal hepatic resection [7]. The safety of cardiac surgery in patients with a cirrhotic liver have been also reported [8]. Suman and coworkers reported that Child–Pugh score and class are useful in predicting postoperative liver failure after cardiac surgery [9]. To our knowledge, there is only one report on lung cancer surgery in cirrhotic patients, by Iwasaki and associates [10]. They reported four liver failure deaths among 17 cirrhotic patients who underwent pulmonary resection for lung cancer.
From the results here, preoperative serum T-BIL was the only predictive factor for postoperative liver failure, though other factors reflecting liver function, such as platelet count, prothrombin test, and ICGR15, also tended to predict this. ICGR15 is a useful marker of liver function that is often used in hepatic surgery for cirrhotic patients [11, 12]. We found the highest ICGR15 values to be 39.5% in a lobectomy case (including mediastinal dissection) and 57.4% in a wedge resection case. In both of these cases, hepatic coma developed transiently in the early postoperative period. Serum AFP was also correlated with postoperative liver failure in our study. AFP is a well-known tumor marker produced by HCC, and has also been reported to reflect the severity of reduced liver function in advanced liver cirrhosis [13, 14].
Bleeding tendency is an important and common complication of liver cirrhosis. In our study, no independent risk factor for postoperative bleeding was found by multivariate analysis. However, the cases with a smaller tumor size or T1 disease had a relatively increased risk of postoperative bleeding. This is probably because, if the pulmonary lesion is sufficiently small, surgeons will perform a standard curative pulmonary resection for lung cancer even in patients with severely diminished liver function, aggressively. In our series, limited pulmonary resection was performed in six cases; four of these were due to reduced pulmonary function, one was due to reduced pulmonary and comorbid with multiple HCC, and one was due to only comorbid multiple HCC. No patients in our series were graded as Child–Pugh class C.
Both deaths in the early postoperative period were due to postoperative critical infectious disease; the sepsis caused acute respiratory distress or multiple organ failure. Neither case had a possible infectious focus or even a low-grade fever preoperatively. As a cause of sudden-onset sepsis in cirrhotic patients without obvious prior infection, bacterial translocation is considerable. Bacterial translocation is ingress of intestinal flora into blood flow, caused by altered permeability of edematous mucosa, which was due to chronically congestive intestinal tract by portal hypertension, and probably was modified by postoperative hemodynamic change [4]. Furthermore, both were noted to have poor nutritional status preoperatively, representing severe abnormality in absorption and metabolism due to decompensatory cirrhosis. Various metabolic abnormalities and malnutrition in cirrhotic patients can induce a serious breakdown of the immune system, especially after major surgical invasiveness [5, 6, 15]. According to our findings, this poor nutritional status is a significant risk factor predicting postoperative critical infection.
In conclusion, patients with cirrhosis can tolerate and could benefit from surgical treatment for lung cancer, while their liver function has been maintained at Child–Pugh class A or B, even with comorbid HCC. Standard lobectomy with mediastinal dissection is favorable if the patient's pulmonary function tolerates. Strict perioperative analysis of the patient's liver function and systemic status, and preparations such as improvement of nutritional status, blood and plasma transfusions, and control of ascites and encephalopathy are important in minimizing peri-operative cirrhosis-related complications.
|
Acknowledgements
|
|---|
We thank Ms Yukiko Wakita for her help in preparing this paper.
|
References
|
|---|
- Tanaka R, Itoshima T, Nagashima H. Follow-up study of 582 liver cirrhosis patients for 26 years in Japan. Liver 1987; 7:316–324.[Medline]
- Llovet JM, Burroughs A, Bruix J. Hepatocellular carcinoma. Lancet 2003; 362:1907–1917.[CrossRef][Medline]
- Amitrano L, Guardascione MA, Brancaccio V, Balzano A. Coagulation disorders in liver disease. Semin Liver Dis 2002; 22:83–96.[CrossRef][Medline]
- Thalheimer U, Triantos CK, Samonakis DN, Patch D, Burroughs AK. Infection, coagulation, and variceal bleeding in cirrhosis. Gut 2005; 54:556–563.[Free Full Text]
- Borzio M, Salerno F, Piantoni L, Cazzaniga M, Angeli P, Bissoli F, Boccia S, Colloredo-Mels G, Corigliano P, Fornaciari G, Marenco G, Pistara R, Salvagnini M, Sangiovanni A. Bacterial infection in patients with advanced cirrhosis: a multicentre prospective study. Dig Liver Dis 2001; 33:41–48.[CrossRef][Medline]
- Vilstrup H. Cirrhosis and bacterial infections. Rom J Gastroenterol 2003; 12:297–302.[Medline]
- Shimada M, Matsumata T, Taketomi A, Shirabe K, Yamamoto K, Itasaka H, Sugimachi K. A new approach for liver surgery. Transdiaphragmatic hepatectomy for cirrhotic patients with hepatocellular carcinoma. Arch Surg 1995; 130:157–160.[Abstract/Free Full Text]
- Lin CH, Lin FY, Wang SS, Yu HY, Hsu RB. Cardiac surgery in patients with liver cirrhosis. Ann Thorac Surg 2005; 79:1551–1554.[Abstract/Free Full Text]
- Suman A, Barnes DS, Zein NN, Levinthal GN, Connor JT, Carey WD. Predicting outcome after cardiac surgery in patients with cirrhosis: a comparison of Child-Pugh and Meld scores. Clin Gastroenterol Hepatol 2004; 2:719–723.[CrossRef][Medline]
- Bizouarn P, Ausseur A, Desseigne P, Le Teurnier Y, Nougarede B, Train M, Michaud JL. Early and late outcome after elective cardiac surgery in patients with cirrhosis. Ann Thorac Surg 1999; 67:1334–1338.[Abstract/Free Full Text]
- Merkel C, Bolognesi M, Finucci GF, Angeli P, Caregaro L, Rondana M, Gatta A. Indocyanine green intrinsic hepatic clearance as a prognostic index of survival in patients with cirrhosis. J Hepatol 1989; 9:16–22.[CrossRef][Medline]
- Hemming AW, Scudamore CH, Shackleton CR, Pudek M, Erb SR. Indocyanine green clearance as a predictor of successful hepatic resection in cirrhotic patients. Am J Surg 1992; 163:515–518.[CrossRef][Medline]
- Furukawa R, Tajima H, Nakata K, Kono K, Muro T, Sato A, Kawahara K, Ishii N, Kusumoto Y, Munehisa T, et al. Clinical significance of serum alpha-fetoprotein in patients with liver cirrhosis. Tumour Biol 1984; 5:327–338.[Medline]
- Jain SK, Rohatgi A, Raman KK, Sharma VK. Study of serum prealbumin and serum alpha fetoprotein in cases of fulminant hepatic failure. J Assoc Physicians India 1995; 43:462–463.[Medline]
- Iida K, Kadota J, Kawakami K, Shirai R, Abe K, Yoshinaga M, Iwashita T, Matsubara Y, Ishimatsu Y, Ohmagari K, Kohno S. Immunological function and nutritional status in patients with hepatocellular carcinoma. Hepatogastroenterology 1999; 46:2476–2482.[Medline]
|
|
Top
Abstract
1. Introduction
