Diabetic ketoacidosis in a case of pheochromocytoma

Diabetes Research and Clinical Practice 54 (2001) 137– 142 www.elsevier.com/locate/diabres

Case report

Diabetic ketoacidosis in a case of pheochromocytoma Chikara Ishii *, Kiyoaki lnoue, Kiyohiko Negishi, Nobuyoshi Tane, Takuya Awata, Shigehiro Katayama The Fourth Department of Medicine, Saitama Medical School, 38 Morohongo, Moroyama-Machi, Iruma-Gun, Saitama 350 -0495, Japan Received 5 July 2000; received in revised form 19 February 2001; accepted 22 March 2001

Abstract A 31-year-old woman was admitted to our hospital because of diabetic ketoacidosis (DKA). Ultrasound sonography revealed the existence of the left adrenal tumor and endocrinological examinations established a diagnosis of pheochromocytoma. She had been healthy and there was no evidence for gestational diabetes in her personal history. Characteristic features were not found in her tumor size and the catecholamine levels as compared with typical cases of pheochromocytoma. An overwhelming secretion of catecholamine might suppress insulin secretion, as evidenced by the improvement after the resection of the tumor. However, a significant insulin resistance continued after tumor resection. Obesity and the heterozygosity of b3-adrenergic receptor gene (Try64Arg) might play a role in insulin resistance, which resulted in DKA at least in part. Literature survey revealed four cases of DKA in the patients with pheochromocytoma including ours, three of which were Japanese. Pancreatic capacity to secrete insulin has been reported to be less than Caucasians, which might be another reason for DKA. Thus, we speculate that both suppressed insulin secretion and insulin resistance deteriorated by obesity or other factor(s) such as abnormality in b3 adrenergic receptor probably depress beta-cell function resulting in abnormal metabolic imbalance such as DKA. © 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Pheochromocytoma; Diabetic ketoacidosis; Insulin secretion; Insulin resistance; Obesity

1. Introduction It has been well known that a mild or moderate impaired glucose tolerance is often associated in the patients with pheochromocytoma (Pheo). However, acute metabolic imbalance such as dia* Corresponding author. Tel.: + 81-492-761204; fax: +81492-949752. E-mail address: [email protected] (C. Ishii).

betic ketoacidosis (DKA) has been thought to be unusual in patients with Pheo. To our best knowledge, only three cases have been previously reported [4–6]. However, the mechanisms or the triggers of DKA in those reported cases were not fully clarified. We report here a case of DKA in the patient with Pheo, and discuss the characteristic features of this case in comparison with the previous reports and typical cases of Pheo without DKA.

0168-8227/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 1 6 8 - 8 2 2 7 ( 0 1 ) 0 0 2 6 1 - 3

138

C. Ishii et al. / Diabetes Research and Clinical Practice 54 (2001) 137–142

2. Case report

Table 1 Clinical data at admission

2.1. Patient

CBC WBC RBC Hb Ht Plt

7500/ml 550×104/ml 16.1 g/dl 48.7% 31.9×104/ml

Biochemistry TP Alb T-Bil D-Bil GOT GPT LDH CPK ALP g-GTP LAP ChE TTT ZTT BUN Cre UA Na K Cl Ca P

7.1 g/dl 4.5 g/dl 0.4 mg/dl 0.1 mg/dl 14 IU/l 14 IU/l 187 IU/l 53 IU/l 346 IU/l 63 IU/l 63 IU/l 6826 IU/l 0.7 KU 0.3 KU 19 mg/dl 0.51 mg/dl 7.8 mg/dl 129 mEq/l 5.2 mEq/l 94 mEq/l 9.2 mg/dl 3.0 mg/dl

Glucose and Lipid metabolism Glucose HbA1c T-Cho TG HDL FFA AcAc 3 OHBA glucagon

433 mg/dl 12.6% 347 mg/dl 232 mg/dl 46 mg/dl 0.90 mEq/l 1400 mmol/l 4400 mmol/l 90 pg/ml

Blood gas analysis (room air) pH PO2 PCO2 HCO− 3 BE

7.273 118.6 mmHg 11.4 mmHg 5.1 mmol/L −20.7

Urinalysis Gravity pH Protein Sugar Ketone body Occult blood Urobilinogen

1.020 5.0 30 mg/dl \ 1.0 g (3+) (+) 0.1

ESR CRP

10 mm/hr 0.16 mg/dl

A 31-year-old woman was admitted to our hospital because of DKA on November 2 in 1996. She had a pregnancy with normal vaginal delivery in 1988. The birth weight of her baby was 3632 g. She had been healthy and there was no evidence of gestational diabetes or other endocrine abnormalities in her personal history. On October 20 in 1995, she suddenly felt flickering and dimming on her eyes and visited a clinic. Hypertension (230/150 mmHg) was pointed out, and hypotensive agents were administered. In April 1996, elevated plasma level of norepinephrine (12 730 pg/ml; normal range 100– 450) was revealed, when she visited a gynecologist at a hospital to consult her sterility. Post-prandial plasma glucose and HbA1c levels were 175 mg/dl and 5.8%, respectively. Although the medication was changed to a1-blocker (urapidil), b1-blocker (metoprolol) and a-methyldopa from June, she was sometimes suffered from severe headache due to paroxysmal hypertension. In October, thirst, polyuria and general fatigue appeared and gradually worsened. She had lost 10 kg in her body weight within a week. On November 2, she admitted to our hospital because of DKA. On physical examination at admission, consciousness was clear. Her height and her body weight were 169.2 cm and 63.5 kg, respectively. Her body temperature, blood pressure and pulses were 37.7°C, 128/80 mmHg and 64 beats per min, respectively. The results of laboratory test at admission are shown in Table 1. No abnormal sign was found on electrocardiogram, a cardiac ultrasonographic examination and X-ray films of the chest and abdomen.

2.2. Clinical course (Fig. 1) After infusion of physiological saline and regular insulin, her acute symptoms of DKA disappeared. Plasma levels of glucose and ketone bodies were maintained near normal levels with basal-bolus insulin therapy. Her blood pressure elevated along with an improvement of dehydra-

C. Ishii et al. / Diabetes Research and Clinical Practice 54 (2001) 137–142

139

Fig. 1. Clinical course after admission. The change of body mass index (BMI), diet energy, therapy including insulin and anti-hypertensive drugs are described in the upper side of the graph. 30R indicates biphasic insulin containing 30% rapid insulin and 70% NPH insulin. R and N indicate rapid insulin and NPH insulin. Solid line and dotted line indicate blood pressure and plasma glucose levels.

tion and an abdominal ultrasonographic examination showed a left adrenal mass. Pheo was diagnosed by the successive examinations. Plasma norepinephrine, epinephrine and dopamine levels were 9676, 193 (normal range; B 100) and 27 (normal range; B20) pg/mi, respectively. Urinary excretion of norepinephrine, epinephrine and dopamine were 4973, 223 and 1244 mg/day, respectively. Their normal ranges were 3– 15, 26– 121 and 190–740. Urinary excretion of metanephrine, vanylmandelic acid and homovanilic acid levels were all elevated, 0.77 (normal range; 0.0.5– 0.23), 24.4 (normal range; 1.5– 6.6) and 5.2 (normal range 1.5–6.6) mg/day, respectively. 131I MIBG adrenal scintigraphy demonstrated an uptake in the left side of her upper abdomen in accordance with the mass described by magnetic resonance imaging (MRI) and digital subtraction angiography. Her blood pressure was controlled within normal range with a1 blocker (doxazosin) and calcium channel antagonist (benidipine). After the resection of left adrenal tumor on day 102, her catecholamine levels and her blood pressure were normalized without medication. Her elevated basic metabolic rate and fasting FFA levels were also normalized. Antibodies against GAD, ICA and insulin were negative. Both plasma glucose

levels and insulinogenic index after 75 g oral glucose load were normalized (Fig. 2). Whereas, insulin sensitivity determined by a hyperinsulinemic euglycemic clamp technique utilizing STG-22 (Nikiso, Tokyo, Japan) revealed that mean clearance rate of glucose was not improved two weeks after operation (3.26 vs. 3.68 mg/kg/min; normal range 6–10 mg/kg/min). The genotype of b3adrenergic receptor gene of our patient was heterozygous (Trp/Arg) at the 64th amino acid residue. The fasting plasma leptin levels in preand post-operation were 3.4 and 2.6 ng/ml, respectively.

Fig. 2. Seventy five gram OGTT before (dotted line) and after (solid line) tumor resection. The left panel represents the change of IRI and right panel represents plasma glucose after glucose load.

C. Ishii et al. / Diabetes Research and Clinical Practice 54 (2001) 137–142

140

Table 2 Clinical data in the present and the previously reported three Pheo cases with DKAa Number (reported)

1 (1992)

2 (1996)

3 (1996)

4 (this case)

Ethnicity Age & Gender Past history Family history BMI (Kg/m2) BP (Max: mmHg) PG (Max: mg/dl) HCO− 3 (mEq/l) Trigger of onset Autoimmunity Tumor mass (cm3) U-Ad (mg/day) U-NAd (mg/day) U-Dp (mg/day) U-VMA (mg/day) U-NME (mg/day) U-ME (mg/day) U-CPR (mg/day) Pre-OP Post-OP insulin resistance

Black (US) 29 F Rheumatic fever NP – 250/150 439 14 Unknown – 7×9.8×8 4701 1798 5890 42 – 6500

Japanese 75 F NIDDM HT (mother) 16.4 134/70 379 8.6 Unknown – 9×6×6 625.0 204.0 – 29.7 – –

Japanese 22 M NP NP 21.1 160/110 494 4.2 Unknown Negative 5×4×3.4 74.4 2656.6 1221.8 10.9 5172.0 538.0

Japanese 31 F NP HT (parents) 22.1 230/150 647 5.1 Unknown Negative 5.5×5×4.5 222.8 4973.2 1244.3 24.4 5370.0 770.0

– – Unknown

38.4 121.6 Unknown

17.0 38.0 Unknown

a

7.3 Positive

NP indicates nothing particular – indicates no data available.

3. Discussion A mild impaired glucose tolerance has been well documented in most cases with Pheo [1– 3]. However, the episode of DKA has been reported in only three cases [4– 6], and the precise mechanisms or features of DKA in patients with Pheo have not been clarified yet. Table 2 shows the clinical data in the present and the previously reported three DKA cases of Pheo [4– 6]. Apparent features were not found in family history, personal history, triggers of DKA, age et al., however depressed insulin secretion by catecholamine was commonly recognized in four cases. In fact, insulin secretion was improved after removal of the adrenal tumor in our case. As regards the suppression of insulin secretion, the effect of hypotensive agents could be neglected as discussed by Edelman et al. [4]. The capacity of insulin secretion, which is known to be lower in Japanese than in Caucasian [7,8], might play an intrinsic role in the development of DKA in three Japanese Pheo cases. While, the type of predomi-

nant catecholamine secreted from the tumor seemed not to affect the development of DKA, because DKA had developed in both epinephrine predominant and norepinephrine predominant cases. Edelman et al. discussed that elevation of epinephrine levels, i.e. more than 400 pg/ml, has been known to have profound effects on glucose intolerance through the a2-mediated inhibitory effects over b2-mediated stimulatory effects of insulin secretion [9–11]. It also induces lypolysis and hepatic gluconeogenesis, and inhibits peripheral glucose utilization [10–12]. While, Cluter et al. reported that the threshold level of plasma epinephrine on lipolytic effect was more than 100 pg/ml, and ketogenic and hyperglycemic effects were 150–200 pg/ml, respectively [10]. The significance of epinephrine levels in our case was not clear, because those levels were around 200 pg/ml. On the other hand, elevated norepinephrine is known to stimulate glycogenolysis in the liver, suppress insulin secretion, increase plasma glucagon concentration, and augment ketone body production from free fatty acids (FFA).

C. Ishii et al. / Diabetes Research and Clinical Practice 54 (2001) 137–142

Furthermore, elevated plasma FFA decreases not only insulin secretion through the reduction of glucose oxidation in pancreatic beta cells, but also insulin sensitivity in the liver and peripheral tissues [13]. However, hyperglycemia has been thought to enhance the re-estirification of FFA in patients with Pheo, which is thought to be reduced ketogenesis [14]. While, in the obese patients like our case, the amount of body fat, especially visceral fat, might play an important role in the development of DKA, because the excessive production of FFA seems to surpass its re-esterification under the significant hyper-catecholaminemic state. Moreover, sustained insulin resistance was observed not only at pre- but also at post-operation state in our case by an euglycemic clamp study. Our patient had been obese until the development of DKA and she also gained 6 kg of body weight within two months after the operation. Her eating behavior and physical activity seemed to be important as the environmental factors of obesity resulting in insulin resistance. Furthermore, thermo genesis related factors should be also important. The role of the genotype of b3-adrenergic receptor gene and TNF-a secreted from visceral fat in insulin resistance has not been identified yet, and it remains to be further clarified. However, resting metabolic rates were reported to be lower in subjects with the mutation of b3-adrenergic receptor gene than in that with normal [15,16]. The genotype of b3-adrenergic receptor gene of our patient was heterozygous (Trp/Arg), which might have worsened her insulin resistance [17,18]. TNF- a level was not measured in this case, however several studies suggest that TNF- a mediates insulin resistance in obese subjects [19– 21]. Leptin has been reported to induce energy expenditure, improvement of insulin sensitivity [22,23], and decrease of triglyceride levels in serum and adipocyte probably through the lipid oxidation [24], in addition to the depression of feeding [25]. In our case, the leptin levels before and after the operation were within normal range, which might not affect insulin resistance. In conclusion, in addition to the suppression of insulin secretion by the sustained hypercatecholaminemia, insulin resistance might deteriorate

141

the beta-cell dysfunction and eventually result in DKA in our case.

References [1] A. De Vries, M. Rachmilewitz, M. Schumert, Pheochromocytoma with diabetes and hypertention, Am. J. Med. 6 (1949) 51 – 59. [2] G.M. Molinatti, F. Masssara, M. Messina, Bilateral pheochromocytoma associated with diabetes, Panminerva Med. 7 (1965) 61 – 66. [3] M. Emmer, F. Gordon, J. Roth, Diabetes in association with other endocrine disorders symposium on diabetes, Med. Clin. North Am. 55 (1971) 1057 – 1064. [4] E.R. Edelman, C.A. Stuenkel, J.D. Rutherford, G.H. Williams, Diabetic ketoacidosis associated with pheochromocytoma, Cleve. Clin. I. Med. 59 (1992) 423 – 427. [5] K. Hashiba, T. Jo, Y. Kazetani, M. Abe, M. Suetsugu, N. Hamada, G. Hiasa, M. Ohmura, A case of diabetic ketoacidosis with pheochromocytoma, Nihon Naika Gakkai-Shi 85 (1996) 1928 – 1930 in Japanese. [6] H. Isotani, Y. Fujimura, K. Furukawa, K. Morita, Diabetic ketoacidosis associated with pheochromocytoma of youth, Diabetes Res. Clin. Pract. 34 (1996) 57 – 60. [7] R. Kawamori, Asymptomatic hyperglycaemia and early atherosclerotic changes, Diabetes Res. Clin. Pract. 40 (1998) S34 – S42. [8] R.H. DeFronzo, The triumvirate. b cells, muscle, liver. A collusion responsible for NIDDM, Diabetes 37 (1998) 667 – 678. [9] W. Manger, R.W. Gifford, Pheochromocytoma, Springer, New York, 1997. [10] W.E. Clutter, D.M. Bier, S.D. Shah, P.E. Cryer, Epinephrine plasma metabolic clearance rate and physiological thresholds for metabolic and hemodynamic actions in man, J. Clin. Invest. 66 (1980) 94 – 101. [11] D. Porte, A receptor mechanism for the inhibition of insulin release by epinephrine in man, J. Clin. Invest. 46 (1967) 86 – 94. [12] S. Ellis, The metabolic effects of epinephrine and related amines, Pharmacol. Rev. 8 (1956) 485 – 562. [13] P.J. Randle, P.B. Garland, C.N. Hales, et al., The glucose fatty acid cycle. Its role in insulin sensitivity and the metabolic disturbance of diabetes mellitus, Lancet 1 (1963) 785 – 789. [14] G. Spergel, S.J. Bleiche, N.H. Ertel, Carbohydrate and fat metabolism in patients with pheochromocytoma, N. Engl. J. Med. 278 (1968) 803 – 809. [15] J. Walston, K. Silver, C. Bogardus, et al., Time of onset of NIDDM and b3-genetic variation in the b 3-adrenergic-receptor gene, N. Engl. J. Med. 333 (1995) 343 – 347. [16] T. Yoshida, N. Sakane, T. Umekawa, et al., Mutation of b 3-adrenergic receptor gene and response to treatment of obesity, Lancet 36 (1995) 1433 – 1434.

142

C. Ishii et al. / Diabetes Research and Clinical Practice 54 (2001) 137–142

[17] E. Widen, M. Lehto, T. Kanninen, et al., Association of a polymorphism in the b 3-adrenergic-receptor gene with features of the insulin resistance syndrome in Finns, N. Engl. J. Med. 333 (1995) 348 –351. [18] N. Sakane, T. Yoshida, T. Umekawa, M. Kondo, Y. Sakai, T. Takahashi, Beta 3-adrenergic-receptor polymorphism: a genetic marker for visceral fat obesity and the insulin resistance syndrome, Diabetologia 40 (1997) 200 – 204. [19] G.S. Hotamisligil, B.M. Spiegelman, Tumor necrosis factor alpha: a key component of the obesity-diabetes link, Diabetes 43 (1994) 1271 –1278. [20] G.S. Hotamisligil, P. Arner, J.F. Caro, R.L. Atkinson, B.M. Spiegelman, Increased adipose tissue expression of tumor necrosis factor-alpha in human obesity and insulin resistance, J. Clin. Invest. 95 (1995) 2409 – 2415.

[21] P.A. Kern, M. Saghizadeh, J.M. Ong, R.J. Bosch, R. Deem, R.B. Simsolo, The expression of tumor necrosis factor in human adipose tissue. Regulation by obesity, weight loss, and relationship to lipoprotein lipase, J. Clin. Invest. 95 (1995) 2111 – 2119. [22] W.I. Sivitz, et al., Effects of leptin on insulin sensitivity in normal rats, Endocrinology 138 (1997) 3395 – 3401. [23] I. Shimomura, et al., Leptin reverses insulin resistance and diabetes mellitus in mice with congenital lipodystrophy, Nature 401 (1999) 73 – 76. [24] G. Chen, et al., Disappearance of body fat in normal rats induced by adenovirus-mediated leptin gene therapy, Proc. Natl. Acad. Sci. USA 93 (1996) 14795 – 14799. [25] M.A. Pelleymounter, et al., Effects of the obese gene product on body weight reduction in ob/ob mice, Science 269 (1995) 540.