InternationalJournalofPancreatotogy,vol. 13,no. 2, 105-110,April 1993 9 Copyright1993by HumanaPressInc. All rights of any naturewhatsoeverreserved.

0169-4197193113:21105-110/$3.00

Free Fatty Acids in Serum of Patients with Acute Necrotizing or Edematous Pancreatitis S. Domschke, P. Malfertheiner, W. Uhl, M. Biichler, and W. Domschke* Departments of Medicine, Universities of Miinster and Ulm, Germany

Abstract Serum concentrations of free fatty acids (FFA) were assayed in 20 patients with acute necrotizing pancreatitis (ANP). Pancreatic and peripancreatic fat necrosis was verified on operation and/or by contrast-enhanced computed tomography. For comparison, 20 patients with acute edematous pancreatitis (AEP) were examined. On admission, FFA serum levels were 1.14 + 0.12 (SEM) mmol/L in ANP and, thus, significantly (p < 0.03) higher than in AEP (0.78 + 0.09 retool/L). The two groups also differed in the later course: in ANP, the FFA values remained raised (d 5-11:0.86 + 0.13 mmol/L; p > 0.05 vs day 1), whereas in AEP, the FFA concentrations normalized within 1 wk (d 2-4:0.52 + 0.11 retool/L; d 5-11:0.39 _+0.05 mmol/L; p < 0.05 vs day 1 and p < 0.0I vs ANP). Serum FFA correlated positively with C-reactive protein levels (r s = 0.42; p < 0.01), but has less discriminating potency between ANP and AEP. In AEP, the initial peak may correspond to the disease outburst itself and to unspecific stress. In ANP, the higher and sustained elevation of FFA may predominantly mirror the ongoing pancreatic parenchymal and extrapancreatic fat necrosis, and be pathophysiologically relevant, especially in view of significantly reduced serum albumin levels in ANP. Key Words: Acute pancreatitis; fat necrosis; free fatty acids; lipolysis; pancreatic necrosis; pancreatic edema; serum calcium; serum triglycerides.

mation (1-3). In recent years, the deleterious contribution of or even a key role for pancreatic lipase has increasingly been acknowledged (4,5). Pancreatic phospholipase itself seems to need lipase cooperation for membrane hydrolysis (6). Experimentally, ip injection of lipase causes massive fat necrosis (5, 7,8), and in vitro, concomitant lipase and olive oil administration decrease the viability of pancreatic acinar cells (9). Serum lipase remains more consistently elevated in acute pancreatitis than amylase (10), and has a higher diagnostic sensitivity and specificity than amylase or trypsin (11). Increased activity of pancreatic lipase, which is already released in active form in the neighborhood (together with colipase), may induce liberation of

Introduction In acute pancreatitis, periacinar release and activation of pancreatic proteases have traditionally been discussed as early and essential pathogenetic steps, although the experimental evidence is scarce (1,2). Likewise, early liberation of phospholipase A 2, which could lead to release of cytotoxic lysolecithin and free fatty acids, has been regarded as an early event in pancreatic and peripancreatic inflamReceivedMay 15, 1992;Revised August24, 1992;Accepted September 1, 1992 *Author to whom all correspondence and reprint requests should be addressed: Department of Medicine B, University of MOnster, Albert-Schweitzer-Str. 33, D-4400 MUnster,Germany 105

106 free fatty acids (FFA) from peripancreatic, retroperitoneal, mesenteric, and omental fat (5). The FFA might, at least in part, reach the general circulation via the portal venous and lymphatic system (5,12). Pancreatitis-associated circulating FFA might thus be a measure of the severity of pancreatitis and, by themselves, contribute to complications in the course of acute pancreatitis, such as the adult respiratory distress syndrome, often associated with hyperlipidemia (13,14}. Locally, FFA might augment the necrotic and inflammatory process, since FFA have been shown to damage the isolated perfused dog pancreas (15-17). Previous studies measuring serum levels of FFA have documented a short-lived (6 h) increase in FFA serum concentrations, which could also be attributed to a likewise short-lived stress-related hormonal imbalance (5,18-20). In the present study, serum FFA were related to the severity of the disease.

Domschke et al. quently during the first 11 d. On the first day, the majority of patients had been fasting for several hours. FFA were assayed by a colorimetric ultramicro method (21). After extracting cobalt soaps of FFA, the complex of Co 2+ with a-nitroso-[3-naphthol was measured at 500 ~m. Until FFA determination, serum samples had been deep-frozen at -70 ~. The following serum levels of parameters relevant in acute pancreatitis were measured by routine methods: C-reactive protein (CRP), albumin, total calcium, and triglycerides. The results of each group and time period were expressed as means + SEM, and the groups of data were compared both by Student's t-test and by the Rank-Wilcoxon test. For the correlation between serum FFA and CRP, the rank correlation coefficient after Speamaan was computed.

Results Free Fatty Acids (FFA)

Materials and Methods Twenty patients with acute necrotizing pancreatitis (ANP), and 20 patients with acute edematous pancreatitis (AEP) with similar age, sex, weight, and aetiologic profile were included into the study. In the ANP patients, pancreatic and peripancreatic fat necrosis was verified by contrast-enhanced computed tomography in all patients. The patients were admitted to the hospital usually within 1-2 d after the onset of their complaints, and presented with the typical clinical signs and symptoms of pancreatitis and serum amylase/lipase values at least twice the upper normal limit. Three patients with ANP died on the third, third, and 59th days (overall mortality 15%). The duration of hospitalization of the other ANP patients was 87 + 9 (SEM) d (p < 0.01 vs AEP). AEP was assessed clinically, by serum lipase and amylase (elevated greater than twice the upper normal limit), and by computed tomography/ ultrasonography. The clinical course was without essential complications in AEP patients. None of the patients died. The duratic, n of hospitalization was 25 + 2 (SEM) d (p < 0.01 vs ANP). For FFA determination, blood samples were obtained on the day of admission (day 1), usually before the therapeutic regimen began, and subseInternational Journal of Pancreatology

On the day of admission, FFA serum levels were 1.14 + 0.12 mmoL/L in ANP and, thus, significantly (p < 0.03) higher than in AEP (0.78 + 0.09 mmol/L). As shown in Fig. 1, the two groups also differed in the later course: in ANP, the FFA values remained raised (days 5-11: 0.86 + 0.13 mmol/L; p > 0.05 vs day 1), whereas in AEP, the FFA concentrations normalized within 1 wk (days 2 - 4 : 0 . 5 2 + 0.11 mmol/L; days 5-11:0.39 + 0.05 mmol/L; p < 0.05 vs day 1 and p < 0.01 vs ANP).

Other Parameters As compiled in Table 1, serum levels of albumin were significantly lower in ANP than in AEP during the whole period of observation. Serum concentrations of C-reactive protein (CRP, Table 1) were elevated both in ANP and AEP (normal values below 5 mg/L), but differed significantly between the two clinical entities. On admission, the CRP values of 19 ANP patients were between 122 and 347 mg/L (only 1 patient had 112 rag/L), whereas the range of 19 AEP patients was 2-113 mg/L (only 1 AEP patient exhibited 319 mg/L). The discriminating potency of CRP was thus much better than that of FFA (Fig. 1). Moreover, CRP values indicated the poor prognosis of the two patients who died after 3 d (347 mg/L in both patients), whereas FFA Volume 13, 1993

Free Fatty Acids and Acute Pancreatitis 2,0

107 250

m

2,5

B

A .J

Ae

.J

o

~

1.5 (/) a

o"

~

<

-2,0

150-

100-

>-

<

50-

UI IX:

L

"-....jr-

t

o

i~'1

i

,

,

o

~

o

i

i

TG 0

0 9

Ca * +

DAY 1

iJ TG

Ca * +

DAYS 2-4

E

-1,5

-1,0

-0,5

TG

Ca*"

DAYS 5-11

IIC W 9

8

o

o

in I

~

i

i

l

i

1

5 NO. OF DAYS IN HOSPITAL

I 11

Fig. 1. Serum levels of free fatty acids in patients with acute necrotizing pancreatitis (A,e), and with acute edematous pancreatitis (A,O) respectively. Means +SEM for day 1, days 2-4, and days 5-11 were calculated. Table 1 Serum Levels (x + SEM) of C-Reactive Protein (CRP, mg/L), and of Albumin (g/L) in Acute Necrotizing Pancreatitis (ANP), and in Acute Edematous Pancreatitis (AEP), Respectively Day 1

Days 2-4

201 + 15a 66 + 16

207 + 24 ~ 42 + 15

109 + 21 ~'b 27 + 15

Albumin ANP AEP

28.2-1- 1.5a 26.1 -t- 2.0 a 25.8 + 1.8a 35.9 + 1.2 35.9 + 1.6 37.6 + 0.8

alndicates p < 0.01 ANP vs AEP. bStands for p < 0.05 vs day 1.

data were less indicative (1.03 and 1.06 mmol/L, respectively; serum albumin was reduced to 23 g/L in both patients). Serum FFA concentrations were, however, positively correlated with plasma levels of CRP (r s = 0.42; p < 0.01). The negative correlation of serum FFA with serum albumin levels (r s = -0.30) just failed to reach the 0 . 0 5 level. International Journal of Pancreatology

As depicted in Fig. 2, serum triglyceride levels were neither increased in ANP nor in AEP; ANP values even tended to be lower than AEP values. Serum calcium was significantly more reduced in ANP than in AEP. Discussion

Days 5-11

CRP ANP AEP

Fig. 2. Serum concentrations of triglycerides (TG) and of calcium (Ca2§ are represented by the darker columns for acute necrotizing pancreatitis (ANP), and by lightly dotted or hatched columns for acute edematous pancreatitis (AEP), respectively. Means +SEM are depicted. The dotted lines indicate the upper normal limit for serum TG and the lower normal limit for serum calcium. *Stands for p < 0.05 when comparing ANP and AEP.

Serum concentrations of FFA seem to reflect the severity of acute pancreatitis. In AEP, the FFA increase in the beginning may correspond to the initial disease itself, to unspecific stress, and to starvation. A transitory FFA elevation of only 6 h has formerly been detected in acute pancreatitis (5,18-20). The rise had been attributed to a likewise limited decrease of serum insulin and an increase in serum levels of lipolytic glucagon and cortisol (5,18-20,22). However, some degree of lipolysis may occur in the pancreas itself, particularly in the compartment of peripheral acinar cells that delivers the enzymes for autodigestive necrosis of neighboring fat cells (4). The inflammatory process with interstitial edema seems to start at these minute necrotic areas (4). Concomitantly with rapid clinical, laboratory, and ultrasonographically demonstrable improvement in AEP, serum FFA rapidly normalized. Volume 13, 1993

108 In ANP, on the contrary, the pancreatitic attack did not subside within 1 wk. The more pronounced and continuous elevation of serum FFA may predominantly mirror the massive and ongoing pancreatic parenchymal anLdcontiguous extrapancreatic fat necrosis. Affected intraabdominal fat is largely converted to FFA, some of which may reach the circulation (5,23). Correspondingly, in experimental necrotizlng pancreatitis, FFA concentrations in ascitic fluid were several times higher than in serum (12). The impoJrtance of fat necrosis for the clinical course is underlined by the fact that obese patients are more endangered by a severe form of acute pancreatitis than nonobese subjects (24,25). In some patients, even distant fat necrosis by circulating lipase may contribute to increased FFA levels (2,5). In addition, lipolytic stress hormones may act on adipose tissue and be, in part, responsible for enhanced FFA liberation (5,18-20,22). Although extrapancreatic necrosis, mainly of fat, is an important risk factor in necrotizing pancreatitis (26), and appeared to be pathophysiologically associated with elevated serum FFA according to our study, the prognostic value of serum FFA is probably limited. The considerable overlap of FFA values between ANP and AEP at the initial presentation of patients allowed no definite conclusion on the outcome of the disease in an individual patient. This is in contrast to the CRP values, which at the usual cutoff limit of 120 mg/L had a sensitivity for detecting ANP of more than 90% in this study, confirming former results (27). Concerning the p~Lthophysiotogical impact, which might be attributed to elevated FFA serum levels in acute pancreatitis: Although FFA are physiological substances and indispensible for energy supply, FFA have to be handled by protective carriers extra- and intracellularly. It has been formerly discussed that FFA may be potentially toxic to proteins and membranes, including the endothelium of the capillaries and the alveocapillary membranes of the lung (13,15,16,28). Pancreatic blood flow can be lowered by FFA (17), and, experimentally, even thrombosis can be induced (29). These bipolar, water--insoluble, detergent-like substances might contribute, for instance, to the development of the adult respiratory distress syn-

International Journal of Pancreatology

Domschke et aL drome with pulmonary edema in severe acute pancreatitis (13,14). FFA toxicity might be augmented by FFA radical formation by other free radicals (30). Normally, FFA are "neutralized" and transported by albumin in the plasma (22,31). The tight, but not covalent, binding of FFA to albumin even enhances hepatic clearance of FFA (31). It has been suggested that in severe pancreatitis, hepatic albumin production may be impaired, especially in concomitant alcoholic liver disease, that albumin degradation may be accelerated (31,32), and that loss of intravascular albumin into the inflamed interstitium, retroperitoneally, and into the peritoneal cavity may lower the serum levels of albumin significantly (2,33), thus increasing the toxicity of FFA (29,34). Correspondingly, serum albumin levels were significantly lower in ANP than in AEP, also in our study. The therapeutic consequence would be first to increase plasma albumin in order to "neutralize" circulating FFA by binding. The therapeutic value of deliberate albumin administration has been made likely in isolated perfused canine pancreas (35), but has not yet been proven in a controlled trial. A recent randomized trial comparing treatment with fresh frozen plasma (2 U daily) with moderate albumin doses revealed no significant difference as to mortality (36). The second approach to decreasing the potential risk of FFA toxicity would be the enhancement of FFA removal from the plasma, e.g., by plasmapheresis (37, 38), and the prevention of supranormal FFA release from adipose tissue or other sources by hypercaloric parenteral alimentation (39,40), insulin substitution (20,22,41--44), and avoidance of lipid infusion in patients with already increased serum levels of FFA (45,46). However, all these therapeutic deliberations have to be assessed in controlled trials.

Acknowledgments Parts of the results have been presented at the Annual 1989 Meetings of the British and German Societies of Gastroenterology and at the World Congress of Gastroenterology, Sydney, Australia, August 29, 1990, and have been published in abstract form (Gut 1989; 30: A742).

Volume 13, 1993

Free Fatty Acids and Acute Pancreatitis

References 1 Keynes M. Heretical thoughts on the pathogenesis of acute pancreatitis. Gut 1988, 29: 1413-1423. 2 Soergel KH. Acute pancreatitis. In: Sleisenger MH, Fordtran JS, eds. Gastrointestinal Disease. Saunders, Philadelphia, 1989. 3 Nevalainen P. Phospholipase A2 in acute pancreatitis. Scand J Gastroenterol 1988, 23: 897-904. 4 K16ppel G, Dreyer Th, Willemer S, Kern HF, Adler G. Human acute pancreatitis: its pathogenesis in the light of immunocytochemical and ultrastructural findings in acinar cells. Virchows Arch (Pathol Anat) 1986, 409: 791-803. 5 Lee PC, Howard JM. Fat necrosis. Surg Gynecol Obstet 1979; 148: 785-789. 6 Mansbach CM. Phospholipases: old enzymes with new meaning. Gastroenterology 1990; 98: 1369-1382. 7 Schmitz-Moormann P, Boger A. Tissue damage by fatty acids released by lipolysis: Contribution in the pathogenesis of acute pancreatitis. Pathol Res Pract 1981; 171 : 303-313. 8 K1OppelG, von Gerkan R, Dreyer Th. Pathomorphology of acute pancreatitis. Analysis of 367 autopsy cases and 3 surgical specimens. Pancreatitis--Concepts and Classification. Gyr KE, Singer MV, Sarles H., eds. Excerpta Medica, Amsterdam, 1984, pp. 29-35. 9 Nagai H, M6ssner J, Wiinsch PH, Henrich H. Ein neues Konzept zur Pathogenese der akuten nekrotisierenden Pankreatitis: Zerstrrung der Zellmembran durch Detergentien? Klin Wochenschr 1988; 66 (Suppl XIII): 157,158. 10 Flamion B, Delhaye M, Horanyi Z, Delange A, Demanet H, Quenon M, Van Melsen A, Cremer M, Delcourt A. Comparison of elastase- 1 with amylase, lipase and trypsinlike immunoreactivity in the diagnosis of acute pancreatitis. Am J Gastroenterol 1987; 82: 532-535. 11 Ventrucci M, Pezzilli R, Gullo L, Plate L, Spovieri G, Barbara L. Role of serum pancreatic enzyme assays in diagnosis of pancreatic disease. Dig Dis Sci. 1989; 34: 39-45. 12 Pappas TN, Gavino VC, Ellison EC, Cornwell DG, Pace WG, Carey LC. Concentration of free fatty acids in pancreatic-associated ascitic fluid. Clin Chem 1981; 27: 358. 13 Kimura T, Toung JK, Margolis S, Bell WR, Cameron JL. Respiratory failure in acute pancreatitis: The role of free fatty acids. Surgery 1980; 87: 509-513. 14 Warshaw AL, Lesser PB, Rie M, Cullen DJ. The pathogenesis of pulmonary edema in acute pancreatitis. Ann Surg 1975; 182: 505-510. 15 Saharia P, Margoli S, Zuidema GD, Cameron JL. Acute pancreatitis with hyperlipemia: Studies with an isolated perfused canine pancreas. Surgery 1977; 82: 60-67. 16 Sanfey H, Bulkley GB, Cameron JL. The pathogenesis of acute pancreatitis. The source and role of oxygen-derived free radicals in three different experimental models. Ann Surg 1985; 201: 633-638. 17 WaldnerH, SchmandJ, VollmarB, Gotz A, ConzenP, Geisler K, Schweiberer L, Brendel U. Acute pancreatitis caused by free fatty acid infusion. Pancreas 1989; 4:647 (A).

International Journal of Pancreatology

109 18 Cameron JL, Capuzzi DM, Zuidema GD, Margolis S. Acute pancreatitis with hyperlipemia: The incidence of lipid abnormalities in acute pancreatitis. Ann Surg 1973; 177: 484--489. 19 Drew S1, Joffe B, Vinik A, Seftel H, Singer F. The first 24 hours of acute pancreatitis. Change in biochemical and endocrine homeostasis in patients with pancreatitis compared with those in control subjects undergoing stress for reasons other than pancreatitis. Am J Med 1978; 64: 795-803. 20 Mizuma K, DeLamater PV, Lee PC, Appert HE, Howard JM. Changing circulating levels of lipids, insulin and glucagon during experimental acute pancreatitis. Surg Gynecol Obstet 1978; 147: 577-582. 21 Nov~kM. Colorimetric ultramicro method for the determination of free fatty acids. J Lipid Res 1965; 6:431-433. 22 Kinney JM, Jeejeebhoy KN, Hill GH, Owen O. Nutrition and Metabolism in Patient Care. Saunders, Philadelphia, 1988. 23 Storck G. Fat necrosis in acute pancreatitis. Acta Chir Scand 1971; Suppl 417: 1. 24 Nordback I, Pessi T, Auvinen O, Autio V. Determination of necrosis in necrotizing pancreatitis. Br J Surg 1985; 72: 225-227. 25 Lankisch PG, Schirren CA. Increased body weight as a prognostic parameter for complications in the course of acute pancreatitis. Pancreas 1990; 5: 626-629. 26 Block S, Btichler M, Bittner R, Beger H. G. Clinical relevance of extrapancreatic necrosis in necrotizing pancreatitis. XX Meeting of the European Pancreatic Club S 70: 1989. 27 Mayer HD, McMahon MJ, Bowen M, Cooper EH. Creactive protein: an aid to assessment and monitoring of acute pancreatitis. J Clin Pathol 1984; 37: 207-211. 28 Wenzel DG, Hale TW. Toxicity of free fatty acids for cultured rat heart muscle and endothelioid cell. I. Saturated long-chain fatty acids. Toxicology 1978; 11: 109-125. 29 Corgeril M, Guidollet J, Renaud S. Serum albumin and mortality. Lancet 1990; 335: 349. 30 Koningsberger JC, Marx JJM, van Hattum L. Free radicals in gastroenterology. Scand J Gastroenterol 1988; 23 (suppl. 154): 30--40. 31 Rothschild MA, Oratz M, Schreiber SS. Serum albumin. Hepatology 1988; 8: 385-401. 32 Streat S, Beddoe AH, Hill GL. Aggressive nutritional support does not prevent protein loss despite fat gain in septic intensive care patients. Aust NZ J Surg 1985; 58: 289-294. 33 Imrie CW, Allam BF, Ferguson JC. Hypocalcaemia of acute pancreatitis: The effect of hypoalbuminaemia. Curr Med Res Opin 1976: 101-116. 34 McMahon MJ. Acute pancreatitis. Curr Opin Gastroenterol 1986; 2: 695-710. 35 Sanfey H, Bulkley GB, Cameron JL. The pathogenesis of acute pancreatitis: the source and role of oxygen-derived free radicals in three different experimental models. Ann Surg 1985, 201: 633-639.

Volume 13, 1993

110 36 Leese T, Holliday M, Heath D, Hall AW, Ball PRF. Multicentre clinical trial of low volume fresh frozen plasma therapy in acute pancreatitis. Br J Surg 1987, 74:907-911. 37 Yamauchi H, Takeda K, Miyagawa K, Sunamura M, Suzuki T, Itoh K, Sato T. Severity and treatment of acute pancreatitis. Nippon Geka Zasshi 1985; 86: 1269-1272. 38 Larvin M, Langdown MRJ, Chalmers AG, Turney JH, Brownjohn AM. Plasmapheresis: a rational treatment for fulminant acute pancreatitis? Lancet 1989; 297: 593-594. 39 Amatruda JM, Salhanick AI. Insulin and steatonecrosis. Hepatology 1989; 10: 1024,1025. 40 Sitzmann JV, Steinborn PA, Zinner MJ, Cameron JL. Total parenteral nutrition and alternative energy substrates in treatment of severe acute pancreatitis. Surg Gynec Obstret 1989; 168: 311-317. 41 Bemal R, Hutson DG, Dombro RS, Livingstone A, Levi JU, Zeppa RA. Possible hepatic factor in the control of plasma free fatty acid levels. Metabolism 1982; 31: 533-537.

International Journal of Pancreatology

Domschke et al. 42 Sauerwein HP, Resola GR, Groeger JS, Jeevanandam M, Brennan MF. Relationship between glucose oxidation and FFA concentration in septic cancer-bearing patients. Metabolism 1988; 37: 1045-1050. 43 Paul P, Issekutz B Jr, Miller HI. Interrelationship of free fatty acids and glucose metabolism in the dog. Am J Physiol 1966; 211: 1313-1320, 44 Shaw JHF, Wolfe RR. Glucose, fatty acid and urea kinetics in patients with severe pancreatitis. The response to substrate infusion and total parenteral nutrition. Ann Surg 1986; 204: 665-672. 45 Akanji AO, Ng L, Humphrey S. Plasma acetate level in response to intravenous fat or glucose/insulin infusions in diabetic and nondiabetic subjects. Clin Chim Acta 1988; 178: 85-94. 46 Ferrannini E, Barrett EJ, Bevilacqua S, De Fronzo RA. Effect of fatty acids on glucose production and utilization in man. J Clin Invest 1983; 72: 1737-1747.

Volume 13, 1993