| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters |
Serv. de
1
Bioquím.,
2
Gastroenterol. and
3
Dermatol., Hospital Severo Ochoa, Avda de Orellana, s/n 28911 Leganés (Madrid), Spain
a Author for correspondence.
To the Editor:
Donnelly et al. (1) described in this journal a
patient similar to one studied by us. Our patient, a 57-year-old man,
presented with cutaneous lesions in his legs, characterized by lobular
panniculitis (PN) with fat necrosis and ghost cells. His serum lipase
was 12 255 U/L at 25 °C (by a turbidimetric method from Boehringer
Mannheim, Mannheim, Germany; upper reference limit, 200 U/L).
Surprisingly, his serum and urine amylase activities were not
increased, and hepatic and renal function appeared to be normal upon
admission. Three weeks later, his serum lipase was 11 141 U/L, and
amylase remained within reference limits. Ultrasonography and computed
tomography (CT) revealed a nodular lesion in his liver (13 x
10 x 15 cm). These features, combined with high lipase activity,
led us to consider a pancreatic carcinoma with liver metastasis.
However, two additional CT scans >2 weeks later showed no pancreatic
abnormality. Histopathological analysis of needle aspirate from a liver
nodule showed a papillary differentiated adenocarcinoma. Intestinal
cancers were searched for but not found. Concentrations of
carcinoembryonic antigen, CA 19.9, and CA 19.5 were within reference
limits. Only 
-fetoprotein was high (1075 µg/L; upper reference
limit: 20 µg/L), consistent with hepatoma (although it can also be
increased less dramatically in pancreatic carcinoma) (2).
Lipase activity was still increased 5 weeks after hospitalization
(24 800 U/L), and activities of hepatic enzymes increased until the
patient's death 3 months after presentation. Two weeks before his
death, ultrasonography revealed a small, hypoechogenic solid node
(2.3 x 2.8 cm) in the body of the pancreas, and the patient was
sent to our reference center for treatment. Postmortem examination
could not be performed.
To account for the high lipase activity, we tested for several possibilities in blood samples obtained from the patient during hospitalization and stored at -20 °C until assay: (a) analytical or physiological effects of drug origin; (b) the presence of macrolipase; and (c) the presence of lipase isoforms (and their pattern). Drug effects (3) were excluded by history. The diagnosis of macrolipasemia was not pursued because <2% of the activity in the supernatant was lost (decreasing from 15 268 to 14 960 U/L) after precipitation with polyethylene glycol as described for macroamylase screening (4). Sera of patients with acute pancreatitis submitted to the same procedure showed a decrease in lipase activity of <20%. A decrease of >73% of amylase activity is generally required before the presence of macroamylase is considered. The precipitation of immunoglobulins after treatment with polyethylene glycol was confirmed by assaying the supernatant for immunoglobulins.
To assay lipase isoforms, we followed the method proposed by Lott and
Lu (5), as modified by Donnelly et al. (1).
For controls, we used sera from patients with acute pancreatitis, in
which L1 and L2 isoforms are usually present in equal amounts
(5). Undiluted serum from our patient showed the presence
of trace amounts of L1 and a large spot at the previously described
L2–L3 position (1)(5) (Fig. 1
). In diluted serum from the patient, a clear band appeared in
the L2 zone. Proximity between L2 and L3 (carboxyl ester lipase) has
been described, but our band is not L3 because the quantitative
turbidimetric method we used does not detect carboxyl ester lipase
activity (6). Accordingly, if only one band appeared in
the L2–L3 zone, it could not be L3. Moreover, in our case, after
incubation for 20 min at 56 °C, the patient's lipase activity
changed from 11 000 U/L to 42 U/L, in agreement with the previously
described thermolability of pancreatic isoforms (7).
|
Our patient presented with an adenocarcinoma associated with high serum lipase activity, similar to the case reported by Donnelly et al. Nowadays, the diagnostic role for lipase isoforms is not clear (6). In their series, Lott and Lu described the presence of L1 and L2 in pancreatic juice and tissue, duodenum extracts, and all acute pancreatitis sera, but not in healthy persons' sera. L2 is also described in sera of patients with gastric, small bowel, or large bowel diseases and in 70% of patients with liver or biliary tract disease, in which the pancreas may be affected by disorders of nearby organs (5). Our unusual serum isoform pattern shows only L2, differing from that described by Donnelly et al., who detected only L1. Lott and Lu described a pattern similar to ours in two patients of their series, but both patients had increased amylase and their lipase was not as high. In the only sample of pancreatic adenocarcinoma tissue they reported, Lott and Lu detected L3 alone, but we do not know the type of adenocarcinoma involved and it showed a low lipase activity of 2.83 U/g of tissue compared with 725 U/g in normal pancreatic tissue. To explain the presence of L1 in serum instead of the L1 and L2 isoforms detected in tumor tissue, Donnelly et al. proposed postranslational modification. The uncertain origin of the primary tumor, with no pancreatic alterations, and the immunohistochemical reactivity of a retroperitoneal node to prostate-specific antigen (PSA) led Donnelly et al. to think that a prostatic carcinoma could be the source of tumor. However, this PSA reactivity has also been described in acinar cell carcinoma of the pancreas (8).
Interestingly, our patient developed PN as a paraneoplastic syndrome, in contrast to the case reported by Donnelly et al. Circulating lipase and other factors may be related to fat necrosis involved in pancreatic PN (9)(10), but not every patient with high serum lipase activities develops this and, as far as we know, no isoform pattern has been studied in tissue and serum of patients with pancreatic PN. Our patient presented with a sort of PN pathognomonic for pancreatic disease (9)(11). This case presents many aspects similar to those described previously (10)(12)(13). The presence of adenocarcinoma, presumably pancreatic, with large hepatic metastases, pancreatic subcutaneous fat necrosis, and very high lipase activities accompanied by normal amylase activity is consistent with acinar adenocarcinoma of the pancreas, an uncommon entity (13) but so aggressive that it can lead to death with disseminated metastasis before its detection in the pancreas. Thus, the biochemical profile may be one of the first findings of this adenocarcinoma.
Acknowledgments
We thank Beckman Instruments (Brea, CA) and Johnson & Johnson Clinical Diagnostic Systems (Rochester, NY), who provided the reagents Paragon isoCK kit and lipase dry-chemistry slides, respectively, for electrophoresis of the lipase isoforms. We also thank Luis Rivas for assistance in the preparation of this manuscript.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  G. Diani, G. Poma, F. Novazzi, S. Zanirato, C. Porta, M. Moroni, G. V. M. d'Eril, and R. Moratti Increased Serum Lipase with Associated Normoamylasemia in Cancer Patients Clin. Chem., May 1, 1998; 44(5): 1043 - 1045. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
