Chest ACCP Education Calendar
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Guest Access | Sign In via User Name/Password
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (11)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Villena, V.
Right arrow Articles by Escribano, P. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Villena, V.
Right arrow Articles by Escribano, P. M.
(Chest. 2002;121:470-474.)
© 2002 American College of Chest Physicians

Amylase Levels in Pleural Effusions*

A Consecutive Unselected Series of 841 Patients

Victoria Villena, MD; Virginia Pérez, MD; Francisco Pozo, MD; Angel López-Encuentra, MD; José Echave-Sustaeta, MD; Joaquín Arenas, MD and Pedro Martín Escribano, MD

* From the Respiratory Medicine (Drs. Villena, Pérez, Pozo, López-Encuentra, Echave-Sustaeta, and Escribano) and Biochemical (Dr. Arenas) Departments, Hospital Universitario 12 de Octubre, Madrid, Spain.

Correspondence to: Victoria Villena, MD, Servicio de Neumología. Hospital 12 de Octubre, Avda. de Córdoba s/ns, 28041 Madrid, Spain; e-mail: mvg01m{at}nacom.es


   Abstract
TOP
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Study objectives: To describe the causes and relative frequency of amylase-rich pleural effusion (ARPE), and to study the origin and histologic type of the tumors with ARPE, the strength of the association between ARPE and the result of pleural cytology, and whether pleural amylase (PA) is a prognostic factor in the survival of patients with a malignant pleural effusion.

Setting: Tertiary-care, university-affiliated hospital.

Patients: Eight hundred forty-one consecutive patients with pleural effusion prospectively assessed from 1991 to 1999.

Results: There were 66 ARPEs: 40 neoplastic, and 26 benign with tuberculosis, pancreatitis, and liver cirrhosis as the most frequent causes. Thirty-six percent of patients in our series and 61% of patients with ARPE had a neoplastic disease (odds ratio [OR], 3; p < 0.001); this association got much stronger for cases with PA levels >= 600 IU/L (95th percentile); [OR, 10; p < 0.001]. The most frequent tumor origin was lung cancer (13 cases). Adenocarcinoma was the most frequent histologic type (18 cases). Two mesothelioma effusions were ARPEs. There was a positive association between ARPE and the finding of tumor cells in pleural fluid (OR, 2.79; p < 0.01). In the malignant group, PA levels >= 600 IU/L identified a group of patients with quite a short median survival (p = 0.016).

Conclusions: The most common cause of ARPE was neoplasm. There was a positive association between ARPE and malignancy, stronger with the highest levels (95th percentile). Lung cancer and adenocarcinoma were the most common tumor and histologic type associated with ARPE. Mesothelioma may also produce ARPE. There was an association between ARPE and the finding of tumor cells in the pleural fluid. The highest PA levels identified a group of patients with a median shorter survival.

Key Words: amylase • amylase-rich pleural effusion • malignant pleural effusion • pleural effusion


   Introduction
TOP
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Pleural effusions with high amylase levels were reported some years ago.1 In addition to pancreatitis and rupture of the esophagus, pleural amylase (PA) levels have been found elevated in patients with malignant and some benign diseases.2 As far as we know, the relative frequency of amylase-rich pleural effusion (ARPE) in different etiologies, or the prognostic value for neoplastic effusions, have not been evaluated in a large unselected series of patients.

The aims of this study are to describe the causes and relative frequency of ARPE; to analyze the histologic type and origin of the tumors with high amylase levels in pleural fluid; to study the strength of the association between amylase levels and the results of pleural cytology; and to know whether PA levels could be a prognostic factor in the survival of patients with malignant pleural effusions.


   Materials and Methods
TOP
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Patients
We prospectively studied 905 consecutive patients with pleural effusions who were assessed on our service from December 1991 to May 1999. The total amylase level in pleural fluid was measured in 841 of these patients (93%), who are our study sample. There were 558 male and 283 female patients. The median age was 67 years (range, 14 to 102 years). All patients were studied according to the same diagnostic algorithm.3 After completion of clinical evaluation, pleural effusions were classified into the following diagnostic groups based on the following explicit criteria:

Malignant: Cytologic or histologic diagnosis of neoplasm within the pleural space.

Paramalignant: Established histologic diagnosis of a tumor in any other organ that was not a malignant effusion, and no other cause of pleural effusion was found.

Tuberculosis: Positive Mycobacterium tuberculosis culture finding in pleural fluid or tissue and/or presence of granulomas in the pleural biopsy specimen, in the absence of other pleural granulomatous diseases. This group also includes 20 patients with high interferon-{gamma} levels in pleural fluid, a favorable clinical course after tuberculous treatment, and either microbiological evidence of extrapleural tuberculosis or a clinical picture suggestive of pleural tuberculosis.4

Parapneumonic Effusion: Presence of cough, fever, and a radiographic pulmonary infiltrate that disappeared with antibiotic treatment.

Transudate: Defined according to clinical and classic biochemical criteria.5

Other Benign Diseases: Diagnosed on the basis of standard criteria.5

Undetermined or Mixed Causes: Unknown effusion etiology or several potential causes of pleural effusion.

Management of Samples
A sample of pleural fluid obtained by thoracentesis and a simultaneous sample of serum were obtained in order to measure the total amylase level. {alpha}-Amylase levels in pleural fluid or serum were measured by the kinetic method based on the cleavage of 4,6-ethylidene-(G7)-1,4-nitrophenyl-{alpha}D-maltoheptaoside by {alpha}-amylase and subsequent hydrolysis of all the degradation products to p-nitrophenol with the aid of {alpha}-glucosidase.6 Both within-run and between-run imprecisions were < 4%. The upper normal serum limit (200 IU/L) was chosen as the cut-off point for defining ARPE.

Statistical Analysis
The Kolmogorov-Smirnov test was used to check the normal distribution of the data. Correlations were studied using the Spearman correlation coefficient. The strength of the associations has been estimated as odds ratio (OR), and {chi}2 test with Yates correction or Fisher’s Exact Test were used as appropriate. A p value < 0.05 was considered statistically significant.

Time to death was calculated from thoracentesis date. The Kaplan-Meier method was used in order to compute the survival probabilities for different categories of PA. The variable was categorized into groups based on the distribution percentiles and compared using the log-rank test.


   Results
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
The etiologies of the effusions from the 841 patients actually included, stratified for pleural fluid amylase levels, are shown in Table 1 . Three hundred one effusions were neoplastic effusions, and 540 effusions were secondary to benign diseases. There were 66 patients with PA levels > 200 IU/L. There was a fair correlation between amylase levels in pleural fluid and serum (r = 0.56; p < 0.001). The distribution of the etiologies of the effusions according to the level of PA is shown in Table 1 . In patients with ARPE, serum amylase was measured in 49 patients, and 43 of these patients had a pleural fluid/serum amylase ratio > 1 (Table 2 ). Twenty-two of these 49 patients had also amylase serum levels > 200 IU/L. In the benign group, three of the patients with ARPE had pancreatitis: one of them had acute pancreatitis, and the other two patients had chronic pancreatitis, both with pancreatic pseudocysts.


View this table:
[in this window]
[in a new window]

  		Table 1.. Etiologies of the Pleural Effusions According to PA Levels

 

View this table:
[in this window]
[in a new window]

  		Table 2.. Higher PA Levels and Pleural Fluid/Serum Amylase Ratio > 1 Within Diagnostic Groups

 
The 11 patients with tuberculosis represent the most frequent benign etiology with ARPE in the present series. Ten patients had pleural granulomas, and 4 of these patients had at least a positive pleural Löwenstein culture finding. The other patient had a very small pleural effusion that precluded a pleural biopsy, a positive induced-sputum finding for tuberculosis, and a pleural interferon-{gamma} level of 25 IU/mL.4

Two patients with parapneumonic effusions had high concentrations of amylase in pleural fluid, one of them with hyperamylasemia with no known pancreatic disease. Of the 100 patients with pleural transudates, only 4 patients, 2 patients with cardiac failure and 2 patients with liver cirrhosis, had ARPEs. One patient with liver cirrhosis reproduced an ARPE 1 year later.

In the group of patients with other benign diseases, only four patients had ARPE: one patient had an uremic pleural effusion, and one patient had a posttraumatic effusion (both 206 IU/L); the third patient was a female with small cell lung cancer, radiation pleuritis, and pericarditis; and the fourth patient had a pleural effusion develop following an abdominal trauma with a splenic hematoma (amylase level, 3,440 IU/L). The high amylase level in this case is likely to be secondary to pancreatic trauma. Finally, in the group of patients with indeterminate or mixed causes, two patients had high amylase levels: one of these patients was a male with lung cancer and pneumonia, and the other patient had gastric cancer and hypoalbuminemia. The pleural effusion was a transudate according to the criteria of Light,5 and the patient died 15 days after thoracentesis. In both cases, the pleural fluid cytology finding was negative.

Thirty-six percent of patients in our overall series, 61% in the ARPE subgroup and 34% in the group without ARPE, had a neoplastic disease, displaying a strong positive association between neoplastic cause of the effusions and ARPE (OR, 3; 95% confidence interval [CI], 1.8 to 5; p < 0.001). This association gets much stronger when we focus on those cases with PA levels >= 600 IU/L (95th percentile); [OR, 10; 95% CI, 2.9 to 34.8; p < 0.001]. Of the 301 neoplastic effusions, 235 effusions (78%) were malignant and 66 effusions (22%) were paramalignant. Forty effusions, 32 malignant effusions (80%) and 8 paramalignant effusions (20%), were ARPEs, as shown in Table 1 .

The distribution of the ARPEs according to the origin and malignant/paramalignant characteristics is given in Table 3 . Table 4 displays information of the relative frequency of ARPE with regard to histologic type and malignant/paramalignant features. We have found that the association between adenocarcinoma and high amylase levels in the neoplastic group is positive and tends to be statistically significant (OR, 1.9; 95% CI, 0.9 to 4.2; p = 0.08).


View this table:
[in this window]
[in a new window]

  		Table 3.. Crossclassification Table of Origin of the Tumors by Malignant/Paramalignant Features and PA Levels*

 

View this table:
[in this window]
[in a new window]

  		Table 4.. Crossclassification Table of Histologic Type of the Tumors by Malignant/Paramalignant Features and PA Levels*

 
In the malignant group, 233 of 235 patients underwent at least one pleural cytology. Nineteen patients (59%) with ARPEs and 69 patients (34%) without ARPEs had positive findings for tumor cells in the first fluid analysis. Accordingly, there is a positive association between high amylase levels and the finding of tumor cells in pleural fluid (OR, 2.79; 95% CI, 1.3 to 6; p < 0.01). After stratifying by histologic type (adenocarcinoma [n = 98], no adenocarcinoma [n = 94], and other histologic types that include undifferentiated carcinoma and unknown [n = 41]), this association is significant only for the adenocarcinoma group (OR, 4.06; 95% CI, 1.2 to 13.4; p = 0.016).

In the malignant group, levels of pleural fluid amylase >= 600 IU/L (95th percentile) identify a small group of patients with quite a short median survival (>= 600 IU/L group, 59 days; < 600 IU/L, 121 days; log-rank test {chi}2, 5.8; p = 0.016). The survival curves are shown in Figure 1 .



View larger version (14K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1.. Survival curves from patients with malignant effusions, according to the amylase level.

 

   Discussion
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
This study describes a large consecutive unselected series of 841 patients in whom levels of PA were measured. To our knowledge, the series published so far have had smaller sample sizes, and the clinical implications of ARPE have not been fully assessed.

We have found ARPE secondary to pancreatic diseases and neoplasms, and commonly associated with a variety of benign etiologies. High levels of amylase in effusions secondary to pancreatitis are widely known. The mechanism seems to be associated with an increase in microvascular permeability.7 Exceedingly high levels have been reported in pancreatic pseudocysts8 ; in fact, two of our patients with this condition reached amylase levels of 65,280 IU/L and 1,839 IU/L, respectively. One of our patients, who presented with acute pancreatitis 2 months before, had an amylase level < 200 IU/L, with no other identified causes of pleural effusion. In this series, no patient had an esophageal rupture.

Pleural tuberculosis has been rarely reported associated with ARPE,2 possibly due to the low prevalence in the studies published so far. Eight percent of our patients with tuberculosis had increased amylase levels, and this etiology represents 17% of ARPEs overall. Although parapneumonic effusions have been found to be a common cause of increased amylase,2 9 we have only found two cases in the 93 patients in this diagnostic group.

As far as transudates are concerned, it has been argued that increased amylase levels in patients with liver cirrhosis could be secondary to some degree of pancreatic inflammation.10 One of our patients with liver cirrhosis and ARPE also had an increased serum amylase level (288 IU/L). Cardiac failure is an uncommon cause of increased pleural fluid amylase.11 Our two patients with this condition had borderline PA and normal serum amylase levels, and neither of them presented clinically significant ascites.

In the other-benign-causes group, three patients with uremia, radiation pleuritis, and abdominal trauma, respectively, had ARPEs. In the undetermined- or mixed-causes group, two patients had ARPEs. Although both of them presented with several factors that might produce pleural effusion and a negative pleural cytology finding, we cannot absolutely rule out malignancy as the cause of those amylase levels.

There is an association between high amylase levels in pleural fluid and a neoplastic cause of the effusion. We agree with other authors2 12 that a very high amylase level strongly suggests a malignant etiology. In fact, in cases with PA levels >= 600 IU/mL, there were 16 with neoplastic effusions (13 malignant and 3 paramalignant), and only 3 benign diseases (2 chronic pancreatitis and 1 abdominal trauma). Nevertheless, being as only 10 to 15% of neoplastic effusions have ARPE, we agree with Branca et al11 that pleural fluid amylase levels should not be routinely measured.

Lung cancer has been reported to be the most common tumor origin of ARPE.2 10 12 13 Our study confirms this finding, but the relative frequency of lung cancer is similar in patients with neoplastic effusions and amylase levels both higher (32%) or lower (34%) than 200 IU/L. The increased rate of lung cancer among the amylase-rich neoplastic fluids might be due to the relative high prevalence of this tumor. Other neoplasms, such as lymphomas2 14 and pancreatic or ovarian carcinomas,2 15 have also been previously reported to produce ARPE.

Adenocarcinoma has been shown to be the most common histologic type among amylase-rich neoplastic effusions.12 13 The trend for a positive association between adenocarcinoma and high amylase levels found in our study tends to confirm the above statement. Two of our 42 patients with mesothelioma had ARPEs, and we agree with Foresti et al12 that high PA levels cannot discriminate mesothelioma from adenocarcinoma, as previously suggested.13

The association between ARPE and the finding of tumor cells in pleural fluid has not been previously reported. Pleural fluid amylase seems to be produced at least in some cases by the tumor cells.1 16 This association might reflect a higher proportion of tumor cells in the pleural fluid.

Finally, we have studied the impact of amylase levels in the survival of patients with malignant effusions. Amylase levels > 600 IU/L were associated with a lower survival, perhaps as a manifestation of more extensive tumor pleural invasion or as a consequence of the tendency of these tumors for lymphatic invasion.10

In conclusion, in our study, the most common cause of ARPE was neoplasm. Of benign diseases, pancreatitis, liver cirrhosis, and tuberculosis were the most common etiologies with ARPE. There was a positive association between high PA levels and malignancy that got stronger when we considered those cases with the highest PA levels. Given its prevalence, lung cancer was the most frequent tumor found in our series. Adenocarcinoma was more commonly associated with ARPE than the other histologic types. Some cases with mesothelioma may produce ARPE. There was a positive association between ARPE and the finding of tumor cells in the pleural fluid that kept only for adenocarcinoma in the stratified analysis. The highest amylase levels identified a group of patients with a median shorter survival.


   Footnotes
 
Abbreviations: ARPE = amylase-rich pleural effusion; CI = confidence interval; OR = odds ratio; PA = pleural amylase

This study has been presented in part at the World Congress on Lung Health and 10th European Respiratory Society Annual Congress, Florence, Italy, August 30–September 2, 2000; and at the XXXII Congreso Nacional del la Sociedad Española de Neumología Cirugía Torácica (SEPAR), Barcelona, Spain, May 15–18, 1999.

Received for publication December 28, 2000. Accepted for publication June 4, 2001.


   References
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Ende, N (1960) Studies of amylase activity in pleural effusions and ascots. Cancer 13,283-287
  2. Joseph, J, Viney, S, Beck, P, et al (1992) A prospective study of amylase-rich pleural effusions with special reference to amylase isoenzyme analysis. Chest 102,1455-1459[Abstract/Free Full Text]
  3. Villena, V, Nieto Barbero, A (1996) Derrame pleural. Martín Escribano P, López Encuentra A, eds ,58-67 Idepsa Pautas de práctica clínica en Neumología. Madrid, Spain.
  4. Villena, V, López-Encuentra, A, Echave-Sustaeta, J, et al (1996) Interferon-{gamma} in 388 immunocompromised and immunocompetent patients for diagnosing pleural tuberculosis. Eur Respir J 9,2635-2639[Abstract]
  5. Light, RW (1995) Clinical manifestations and useful test. Light, RW eds. Pleural diseases 3rd ed. ,36-74 Williams & Wilkins Baltimore, MD.
  6. Lorent, K (1998) Approved recommendation on IFCC methods for the measurement of catalytic concentration of enzymes: part 9. IFCC method for {alpha}-amylase (1,4-{alpha}-D-glucan 4-glucanohydrolase, EC 3.2.1.1.) Clin Chem Lab Med 36,185-203[CrossRef][ISI][Medline]
  7. Joseph, J, Smith, MJ, Basran, GS (1990) A study of pleural effusion kinetics in man using a dual isotope method [abstract]. Chest 98,138S
  8. Lorch, DG, Sahn, SA (1987) Pleural effusions due to diseases below the diaphragm. Semin Respir Med 9,75-85
  9. Light, RW, Ball, WC (1973) Glucose and amylase in pleural effusions. JAMA 225,257-260[CrossRef][ISI][Medline]
  10. Ende, N (1961) Amylase activity in body fluids. Cancer 14,1109-1114
  11. Branca, P, Rodriguez, RM, Royers, JT, et al (2001) Routine measurement of pleural fluid amylase is not indicated. Arch Intern Med 161,228-232[Abstract/Free Full Text]
  12. Foresti, V, Villa, A, Zubani, R (1994) Amylase concentrations in pleural effusions. Chest 105,1625-1626[ISI][Medline]
  13. Kramer, MR, Saldana, MJ, Cepero, RJ, et al (1989) High amylase levels in neoplasm-related pleural effusion. Ann Intern Med 110,567-569
  14. Zimmerman, HM, Bank, S, Bouch, P, et al (1983) Macroamylase in pleural fluid of a patient with lymphoma. Gastroenterology 85,190-193[Medline]
  15. Cramer, SF, Burns, DF (1979) Amylase producing ovary neoplasm with pseudomeigs syndrome and elevated pleural fluid amylase. Cancer 44,1715-1721[Medline]
  16. Devuyst, O, Lambert, M, Scheiff, JM, et al (1990) High amylase activity in pleural fluid and primary bronchogenic adenocarcinoma. Eur Respir J 3,1217-1220[Abstract]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (11)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Villena, V.
Right arrow Articles by Escribano, P. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Villena, V.
Right arrow Articles by Escribano, P. M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS