Pancreas n DM

8/9/2019 Pancreas n DM

1/10

Pancreatic Diabetes Mellitus

Robert |. Sjoberg, MD

Gerald S. Kidd, MD

Diabetes mellitus caused by pancreatic exocrine disease

is a unique clinical and metabolic form of diabetes. The

diagnosis of pancreatic diabetes caused by chronic

pancreatitis may be elusive because it is occasionally

painless and often not accompanied by clinical

malabsorption until after hyperglycemia occurs. Diabetic

patients with pancreatic calcification or clinically

demonstrable pancreatic exocrine dysfunction will

manifest the unique aspects of pancreatic diabetes

described herein. Like other forms of diabetes, the

primary hormonal abnormality in pancreatic diabetes is

decreased insulin secretion. Patients with this disorder

are unique in that they have low glucagon levels that

respond abnormally to several physiological stimuli,

blunted epinephrine responses to insulin-induced

hypoglycemia, and malabsorption. In addition, they

often have concomitant alcohol abuse with hepatic

disease and poor nutrition. These characteristics result

in increased levels of circulating gluconeogenic amino

acids, decreased insulin requirements, a resistance to

ketosis, low cholesterol levels, an increased risk of

hypoglycemia w hile on insulin therapy, and the clinical

impression of brittle diabetes. Retinopathy occurs at a

rate equal to that of insulin-dependent diabetes but

may be less severe in degree. Other complications of

pancreatic diabetes have been less well studied but may

be expected to be seen more frequently as these

patients survive longer. The characteristics of pancreatic

diabetes suggest that a conservative approach be taken

in regard to intensive insulin therapy and tight blood

glucose con trol.Diabetes Care12:715-2 4, 1989

From the Endocrine Service, Department of Medicine, Fitzsimons Army Medical

Center, Aurora, Colorado.

Address correspondence and reprint requests to Gerald S.

Kidd,

MD, Endo-

crine Service, Department of Medicine, Fitzsimons Army Medical Center, Au-

rora,

CO

80045-5001.

The opinions and assertions contained herein are the private views of the

authors and are not to be construed as official or reflecting the views of the

Department of the Army or the Department of Defense.

A

lthough the exocrine and endocrine cells of the

pancreas are generally thought to function in-

dependently of each other, they are anatomi-

cally closely related. This anatomic relationship

allows pancreatic islet cell dysfunction to be associated

with and/o r caused by pancreatic exocrine disease. The

pancreatic exocrine syndromes known to be associated

with so-called pancreatic diabetes include acute and

chronic pancreatitis, postpancreatectomy syndrome,

cystic fibrosis, tropical pancreatic diabetes, and pan-

creatic carcinoma. This association is historically im-

portant because the role of the pancreas in causing di-

abetes mellitus was initially postulated 200 yr ago in a

report of a patient with diabetes who had pancreatic

calcifica tion (1). Pancreatic diabetes continues to be im-

portant because it is unique clinically, hormonally, and

metabolically, prompting the National Diabetes Data

Group to classify it as a distinct form of diabetes (2).

This review summarizes the clinical manifestations of

pancreatic diabetes, with particular emphasis on pan-

creatitis and postpancreatectomy syndrome. Diabetes

associated with cystic fibrosis and tropical pancreatic

diabetes is discussed briefly in comparison with other

forms of pancreatic diabetes, but it has recently been

reviewed in greater detail (3-5).

FREQUENCY

AND

DIAG NO SIS

Mild hyperglycemia may occur transiently in association

with acute pancreatitis up to 50% of the time and has

been used as a marker for the severity of pancreatitis

(6,7). Long-term follow-up of patients after a single ep-

isode of acute pancreatitis indicated that they had an

increased frequency of abnormal glucose tolerance

compared with an age-matched population (8). Such

DIABETES CARE, VOL. 12, NO . 10, NOVEMBER/DECEMBER 1989 715


2/10

PANCREATIC DIABETES MELLITUS

patients may have chronic painless pancreatic inflam-

mation in addition to their acute painful episode (6). A

distant history of acute pancreatitis may, however, be

etiologically important in any patient with diabetes.

The prevalence of abnormal glucose tolerance in pa-

tients with chronic pancreatitis was 60-70% and of overt

diabetes was 3 0- 40 % (6,9,10). Serial testing of individ -

ual patients with chronic pancreatitis demonstrated a

progressive deterioration of glucose tolerance over

time (6).

The presence of pancreatic calcifica tion increased the

occurrence of endocrine dysfunction, resulting in a 70%

prevalence of diabetes (6). Thirty-three percent of pa-

tients with chronic pancreatitis and diabetes had pan-

creatic calcification (6). Pancreatic calcification was

present in 20% of patients with chronic pancreatitis

but was more common in alcohol-induced (30%) and

tropical (75%) pancreatitis (4-6).

Chronic pancreatitis is painless 5-10% of the time,

and steatorrhea is a late manifestation of chronic pan-

creatitis, often occurring after the onset of diabetes

(6,9-11). This implies that the diagnosis of diabetes as

a result of chronic painless pancreatitis can be elusive.

Approximately 30% of patients with diabetes and cal-

cific pancreatitis in one large series were diagnosed by

abdominal roentgenography of patients previously thought

to have genetic diabetes (6). The data reported below

regarding the clinical and hormonal status of patients

with chronic pancreatitis were derived from patients with

either pancreatic calcification on plain roentgenography

or well-documented recurrent episodes of pancreatitis

associated with pancreatic exocrine insufficiency and

usually a known etiology for chronic pancreatitis. With

these points in mind, proposals regarding who in the

diabetic popu lation should be screened for and practical

clinical criteria for the diagnosis of pancreatic diabetes

are given in Table 1. Although these criteria may not be

inclusive, we feel that patients meeting these criteria w ill

manifest the unique aspects of pancreatic diabetes out-

lined below.

Little is known regarding the cause of chronic pan-

creatitis and any effect this may have on the diabetic

syndrome. Most patients studied have had alcohol abuse

as the cause for their pancreatic disease. This may sim-

ply reflect the most common etiology of chronic pan-

TABLE 1

Clues and criteria for diagnosis of pancreatic diabetes

Clues

Suspect pancreatic diabetes in any diabetic patient with:

One or more d ocumented episodes of acute pancreatitis

A known historical cause for chronic pancreatitis

Previous pancreatic surgery

Criteria

Patients have pancreatic diabetes if:

Pancreatic calcification is present on plain roentgenography

Steatorrhea partially reversible with pancreatic enzyme therapy

is demonstrable

creatitis, but it has also been suggested that alcohol-

induced pancreatitis results in a higher prevalence of

diabetes than other etiologies of pancreatitis (6,9,11).

Cystic fibrosis results in diabetes less frequently than

chronic pancreatitis, its prevalence being 7-15% of cys-

tic fibrosis patients >18 yr of age (3). Tropical pan-

creatic diabetes differs from other causes of pancreatic

diabetes in its higher insulin requirements (4,5). Total

pancreatectomy generally results in a more severe de-

gree of endocrine dysfunction than chronic pancreatitis,

but resection of


3/10

R.J. SJOBERG AN D G.S.KIDD

1050

9 0 0

^ 750

40,000, but low levels of 3500-M

r

glucagon were

occasionally detectable in such patients (31-33). The

paradoxical increase in immunoreactive glucagon after

oral glucose ingestion was also caused by an increase

+80

+

60

+

40

+20

- 2 0

1 4 0 r B

120

\

-1

1 0 0

I

8 0

6 0

3

2 0

0

L

5- -S--


4/10

PANCREATIC DIABETES MELLITUS

in a 9000- to 15 ,000-M

r

fraction of immunoreactive glu-

cagon (34).

The source of the low levels of 3500-M

r

glucagon in

pancreatectomized humans is unclear. All pancreatec-

tomized patients studied have also had a partial duo-

denectomy and often an antrectomy, ruling out these

tissues as potential glucagon sources. The possibility of

peripheral conversion of gut-derived high-molecular-

weight glucagon to 35OO-M

r

glucagon must also be con -

sidered (35).

AMINO ACID METABOLISM

Plasma amino acids, especially those that serve as

substrates for gluconeogenesis, were elevated postpan-

createctomy compared with normal or IDDM levels

(31,32,36-39). Maintenance of normoglycemia for 24 h

with a glucose-controlled insulin infusion system did not

decrease these gluconeogenic precursors to normal as

it did in IDDM (38). A physiological glucagon infusion

did, however, normalize amino acid levels, suggesting

that hypoglucagonemia mediates hyperaminoacidemia

(32,37,39).

FATTY ACID AND KETONE BODY METABOLISM

Studies concerning fatty acid metabolism postpancrea-

tectomy helped clarify the role of glucagon in the de-

velopment of ketoacidosis. Withdrawal of insulin from

fasting pancreatectomized patients resulted in ketoaci-

dosis, but blood ketones and glucose rise less markedly

over time than in IDDM patients (40,41; Fig. 3). No

difference was found between these two types of dia-

betes in free fatty acid increments (40). IDDM patients,

unlike pancreatectomized patients, had a marked in-

crease in glucagon concentration during development

of ketoacidosis (40; Fig. 4). These data suggest that

hyperglucagonemia promotes hepatic conversion of free

fatty acids to ketone bodies, but in view of the detect-

able plasma levels of 3500-M

r

glucagon found in some

pancreatectomized patients, the absolute need for glu-

cagon in the development of ketoacidosis remains un-

15-i STOP

INSULIN

2 10

o

o

o

5 -

0

J

0 -

i i

-1 0

4 8

TIME (hours)

12

- 1 0

4 8

TIME (hours)

12

FIG. 3. Changes in blood concentrations of 3-hydroxybutyrate and glucose in 6 patients with insulin-depende nt dia-

betes () and 4 pancreatectomized patients (O) after withdrawal of insulin. Values are means SE.*P


5/10

R.J. SJOBERG AND G.S. KIDD

30

1

o

o

Documents

Pancreas n DM