- Email: [email protected]
Recurrent unexplained hyperammonemia in an adolescent with arginase deficiency
Clinical Biochemistry 45 (2012) 1583–1586
Contents lists available at SciVerse ScienceDirect
Clinical Biochemistry journal homepage: www.elsevier.com/locate/clinbiochem
Recurrent unexplained hyperammonemia in an adolescent with arginase deficiency Yan Zhang a,⁎, Yuval E. Landau b, David T. Miller b, c, d, Deborah Marsden b, d, Gerard T. Berry b, d, Mark D. Kellogg c, e a
Department of Pathology and Laboratory Medicine, University of Rochester Medical Center, Rochester, NY, USA Division of Genetics, Children's Hospital Boston, Boston, MA, USA Department of Laboratory Medicine, Children's Hospital Boston, Boston, MA, USA d Department of Pediatrics, Harvard Medical School, Boston, MA, USA e Department of Pathology, Harvard Medical School, Boston, MA, USA b c
a r t i c l e
i n f o
Article history: Received 29 May 2012 Received in revised form 10 August 2012 Accepted 12 August 2012 Available online 23 August 2012 Keywords: Hyperammonemia Pre-analytical variations Arginase deficiency Urea cycle disorders
a b s t r a c t Objectives: This report investigates the etiology of recurrent episodic elevations in plasma ammonia in an adolescent male with arginase deficiency as there were concerns regarding pre-analytical and analytical perturbations of ammonia measurements. There were repeated discrepancies between the magnitude of his ammonia levels and the severity of his clinical signs of hyperammonemia. Patient and methods: The patient is a fourteen-year-old arginase-deficient male diagnosed at three years of age. Since 2008 (when he reached 10 years of age), there appeared to be an increase in the frequency of hospitalizations with elevated ammonia. A typical emergency visit with initial ammonia of 105 μmol/L (reference interval: 16–47 μmol/L) is illustrated. Pre-analytical and analytical procedures for the patient's sample handling were retrospectively examined. His ammonia levels were compiled since diagnosis. The frequency of his initial or peak ammonia levels greater than two times (94 μmol/L) or four times (188 μmol/L) the upper limit of normal was computed. Student t-test was used to calculate the significance of the differences before 2008 and since 2008. Results: One out of eleven and ten out of 19 hospitalizations had initial ammonia greater than two times normal before and after 2008, respectively. Both the patient’s overall ammonia and peak ammonia levels are significantly higher since 2008 (p value b 0.001 for both) than those before 2008. Conclusions: To our knowledge, few adolescent males with arginase deficiency experience recurrent episodes of hyperammonemia requiring intravenous nitrogen scavenging agents. We hope that this study provides new insights into the natural history of arginase deficiency and the management of such patients. © 2012 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Introduction The urea cycle consists of six consecutive enzymatic reactions to remove excess ammonia that is generated during the catabolism of nitrogen-containing compounds from the body [1]. Ammonia is in equilibrium with glutamine and is converted to urea through the urea cycle. Plasma ammonia concentrations are maintained within a fairly narrow range (reference interval [RI]: 16–47 μmol/L). Elevations are usually associated with acquired liver disease such as liver cirrhosis or severe hepatitis, or with certain organic acid disorders (propionic acidemia) or urea cycle disorders. Hyperammonemia, with one or several clinical signs of lethargy, confusion, irritability, vomiting, seizure, tremor, and coma, is a medical emergency requiring immediate medical
⁎ Corresponding author at: Dept. of Pathology and Laboratory Medicine, University of Rochester Medical Center, 601 Elmwood Avenue Box 608, Rochester, NY 14642, USA. Fax: +1 585 273 3003. E-mail address: [email protected] (Y. Zhang).
attention [1]. Prolonged exposure to high ammonia can lead to severe central nervous system damage and even death [1]. Arginase deficiency (also known as hyperargininemia, Mendelian Inheritance in Man number 207800), is caused by a deficiency of liver enzyme arginase I (E.C. 3.5.3.1). It is the least common urea cycle disorder with an estimated prevalence of 1 in 1,100,000 [2]. Arginase, the last enzyme in the urea cycle, catalyzes the conversion of arginine to urea and ornithine (Fig. 1). Urea is excreted in the urine, and ornithine is transported to the mitochondria to continue the cycle. Clinical manifestations of arginase deficiency are strikingly different from the other urea cycle disorders. As opposed to other urea cycle disorders that may present 24 to 72 h after birth or occasionally several days later into the extended neonatal period, arginase deficiency rarely presents in the neonatal period and first symptoms typically present in children between two and four years of age [3]. Hyperammonemia is not typically associated with arginase deficiency [4]. When hyperammonemic encephalopathy occurs, the plasma ammonia levels are three to four times normal value, with levels rarely greater than six times normal. These patients can become comatose
0009-9120/$ – see front matter © 2012 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.clinbiochem.2012.08.015
1584
Y. Zhang et al. / Clinical Biochemistry 45 (2012) 1583–1586
Fig. 1. Schematic of the urea cycle. Arginase is the last enzyme in this cycle which converts arginine to ornithine and urea.
when their ammonia levels rise to three to four times normal and may succumb to lethal brain edema [5]. Although not well studied, elevated arginase II, the mitochondria form of arginase expressed in kidney and brain, has been reported to be further induced by increased arginine levels in patients with arginase deficiency [6,7]. Arginase II may be the major contributor to the generally milder clinical presentations in these patients in comparison to other urea cycle disorders [5]. The less severe clinical course for arginase deficiency may also be due to the fact that arginase is at the last step of the urea cycle and two molecules of ammonium ions have already been accumulated into L-arginine by this point. In this report, we describe a male with arginase deficiency who had multiple episodes of hyperammonemia that required intravenous nitrogen scavenging medications between 10 and 14 years of age. In the majority of these episodes, he demonstrated less severe clinical presentation than expected based on the high ammonia levels. One typical emergency hospital visit is illustrated here during which he did not show typical clinical signs of hyperammonemia even at ammonia levels of 252 μmol/L and 242 μmol/L. This significant discrepancy triggered an investigation in the laboratory to identify 1) whether the ammonia level was falsely elevated due to pre-analytical and/or analytical problems, or 2) whether the discrepancy was related to the patient's underlying arginase deficiency.
However, his baseline is that of severe developmental disability, behavior problems and spastic diparesis. He has had 19 hospitalizations with elevated ammonia (>47 μmol/L) since 2008 with most of them requiring hydration and Ammonul (see description below) therapy. Eleven of these visits showed peak ammonia levels of more than four times normal (188 μmol/L). The average peak ammonia was 224 μmol/L (standard deviation [SD]=127 μmol/L). In the majority of these episodes, the clinical signs of hyperammonemia did not match the severity of his elevated plasma ammonia. In comparison, he had 11 hospital visits due to elevated ammonia between 2000 and 2008 and only three visits showed peak ammonia levels of two times normal (94 μmol/L) and none beyond four times normal (Fig. 2B). During this period, the average of his peak ammonia was 80 μmol/L with SD of 31 μmol/L. One of his typical emergency visits is described below. The patient presented to the emergency room with lethargy, emesis, and parental concerns about an “impending” metabolic crisis. The patient had complained of increased fatigue for several days, according to his mother. He had refused to eat by mouth on the day of admission and had been given hypercaloric and protein-free formula through his G-tube. Laboratory data obtained upon arrival for this visit indicated his ammonia was 105 μmol/L. Later analysis of the admission sample found arginine was 384 μmol/L (RI: 31–124 μmol/L) and decreased values for threonine, serine, isoleucine, and leucine. Glutamic acid and glutamine were within the reference intervals and gamma-glutamyltransferase (GTT) was 16 IU/L
Case description The patient is a fourteen-year-old male with arginase deficiency diagnosed at three years of age based on reduced arginase activity in red blood cells (November 2000). The patient had a history of severe intellectual disability, seizure disorder, asthma, gastrointestinal complications, and spastic diparesis. The patient was put on a protein-restricted diet (December 2000) soon after diagnosis. His most recent diet is composed of 0.4 g/kg of essential amino acids from medical food (formula), and 0.4 g/kg of intact protein from food. If he doesn't eat his full-recommended amount of protein, it is given as Lactaid milk through the gastrostomy tube (G-tube). The mother keeps detailed food records and typical foods include Cheese Nips, Cheez-Its, Wise onion rings, Light & Lively yogurt, Pringles, low protein cheese, and animal crackers. His height (by the age of 14) was 160.3 cm (29th percentile for age, z-score 0.55), weight is 63.6 kg (85th percentile for age, z-score 1.02), and body mass index is 24.8 kg/m2 (93rd percentile for age, z-score 1.44). Medications included sodium phenybutyrate (13.5 g/day), sodium benzoate (8.1 g/day), clonidine, speridone, Topamax, clinazepam, trazadine, Miralax, and Prilosec. The patient is generally well when there is no hyperammonemia.
Fig. 2. A. All ammonia measurements from the patient between 2000 and 2011. The patient was diagnosed with arginase deficiency in 2000 nearly at age of three. The p value indicated the difference of ammonia measurements before and after 2008. B. The initial and peak ammonia levels for patient's hospital visits due to elevated ammonia. The open squares are for peak ammonia and the open diamonds are for initial ammonia. The p value indicated the difference of peak ammonia before and after 2008.
Y. Zhang et al. / Clinical Biochemistry 45 (2012) 1583–1586
(RI: 12–55 IU/L). Hemolysis, icterus and lipemia indices for the sample were all below levels expected to cause interference. The patient was immediately evaluated in the emergency department and admitted to the medical intensive care unit for management of hyperammonemia. He was administered a continuous intravenous infusion of Ammonul (sodium benzoate/sodium phenylacetate; Medicis Pharmaceuticals Corp., Scottsdale, AZ), each at 5.5 g/m2/day in 10% glucose. An additional intravenous infusion of 20 mEq/L KCl was given so that the total fluid rate was 1.5 times maintenance. The sodium phenylbutyrate and sodium benzoate medication, given orally on a daily basis, were stopped. All other outpatient medications were continued. The components of Ammonul serve to enhance waste nitrogen excretion through the production of alternate waste nitrogen vehicles in the treatment of urea cycle disorder associated hyperammonemia. Specifically, phenylacetate and sodium benzoate are conjugated with glutamine and glycine, respectively, to form phenylacetylglutamine and hippuric acid which are then excreted through the kidneys [8,9]. Because the patient's plasma ammonia was only 105 μmol/L and the clinical signs of hyperammonemia were mild, it was decided to not administer the usually prescribed bolus dose of Ammonul (5.5 g/m 2/90 min). The therapy might also start at a lower dose based on the clinical situation. In the meantime, he was given ProPhree® (Abbott Nutrition, Columbus, OH), a protein-free formula, through the G-tube for additional calories. The protein-free diet was given for no more than 24–48 h after which he could start gradually getting his regular mix of essential amino acid formula with whole protein food. His plasma ammonia levels were closely monitored. The patient's general condition as well as his neurologic status appeared to improve shortly after admission and the lethargy did not worsen. The patient was not comatose, and he was not manifesting hyperventilation or seizure activity or any exacerbation in upper motor neuron signs such as the emergence of spontaneous ankle clonus. However, his ammonia level reached 252 μmol/L from a venous sample. An arterial blood sample obtained approximately 90 min after the venous sample measured an ammonia level of 242 μmol/L. These elevated ammonia levels did not reflect his improved clinical presentation. The following morning, the patient's lethargy had resolved, and the ammonia level had decreased to 65 μmol/L. It continued at about the same level for the next 60 h (Fig. 3).
Results and discussions Pre-analytical and analytical issues of ammonia measurement Ammonia is known to be one of the most difficult analytes to measure. Questions were raised for our patient since hyperammonemia is
Fig. 3. Plasma ammonia levels after patient's admission. Ammonia levels over a course of 60 h are shown. Dotted line, the reference interval for males, 16–47 μmol/L. Solid line, critical value in our hospital, 58 μmol/L.
1585
not typically associated with arginase deficiency patients and his clinical presentations did not match his elevated ammonia levels. The accuracy of ammonia measurements can be affected by either pre-analytical or analytical phases or both with the former being the major source of spuriously elevated ammonia results [10]. Delayed processing and hemolysis can lead to falsely elevated ammonia. Ammonia concentration may also be increased in vitro from the decomposition of glutamine seen in patients with high GTT activity [11], and elevated glutamine levels have also been reported in patients with urea cycle disorders, particularly patients with late-onset presentation [12]. The goal for reliable ammonia measurement is to minimize any release of ammonia from collected blood samples before analysis. The consensus statement from the conference on the management of patients with urea cycle disorders suggests that non-hemolyzed plasma samples should be collected in a pre-chilled, ammonia-free EDTA tube, kept on ice, and should be centrifuged within 15 min of collection [13]. Interpretation of ammonia results from patients with metabolic or liver pathology is significantly more challenging since samples from these patients can contain several sources of ammonia including ammonia formed in vitro at various rates and amounts. The analytical concerns associated with ammonia measurement by titration, colorimetric, fluorimetric, or gas-sensing electrode methodologies [14] have been largely addressed by the introduction of enzymatic assays based on glutamate dehydrogenase. In this report, the ammonia was measured on a Roche Cobas Integra800 instrument. The glutamate dehydrogenase specifically catalyzes the reduction of 2-oxoglutarate with NH4+ and NADPH to form L-glutamate and NADP. The concentration of NADPH consumed, which is determined by measuring the decrease in absorbance at 340 nm, is directly proportional to the ammonia concentration. For our patient, it was confirmed that all samples were collected in pre-chilled EDTA collection tubes and delivered on ice. Restrictions on tourniquet use, fist clenching and muscle activity were observed by staff performing blood collections. The samples were processed promptly, and the results were reported within 30 min. While the exact causes remain unknown, there were no obvious analytical or pre-analytical concerns for the elevated ammonia levels in this patient. The clinical features of arginase deficiency Arginase deficiency patients do not typically present with hyperammonemia in comparison to those with other urea cycle disorders. It's even rarer to see episodic hyperammonemia requiring intravenous Ammonul treatments in these patients. Our patient has had more severe ammonia elevations and increased frequency of hospital visits due to hyperammonemia since he entered the second decade of his life (2008) (Fig. 2A/B). The patient's ammonia levels have been significantly higher since 2008 (p valueb 0.001, N=89 before 2008, N=220 after 2008, Fig. 2A). The mean ammonia levels were 42 (SD=26) and 95 (SD=79) μmol/L for all the available ammonia measurements before and after 2008, respectively. Before 2008, there were no ammonia measurements beyond four times normal (maximum=131 μmol/L), while after 2008, 26 ammonia measurements have been recorded beyond that limit with six of them showing even more than 300 μmol/L (maximum=477 μmol/L) (Fig. 2A). Of his 19 hospital visits due to hyperammonemia since 2008, ten of them had initial ammonia more than two times normal and three visits had initial ammonia more than four times normal (Fig. 2B). Before 2008, however, only one hospital visit out of eleven showed an initial abnormal ammonia level of two times normal (Fig. 2B). The peak ammonia levels were also significantly higher after 2008 (mean = 224 μmol/L, SD = 127 μmol/L) than before 2008 (mean = 80 μmol/L, SD = 31 μmol/L, p b 0.001). Sixteen visits since 2008 showed peak ammonia levels more than two times normal and eleven visits more than four times normal while only three visits before
1586
Y. Zhang et al. / Clinical Biochemistry 45 (2012) 1583–1586
2008 had peak ammonia levels more than two times normal and none were four times normal or greater (Fig. 2B). His typical treatments during hyperammonemia involved intravenous ammonia reduction agents such as Ammonul as indicated in the Case description section. The patient was well when his ammonia was under control. Episodic hyperammonemia is well known for other types of urea cycle disorders. To our knowledge, there are few adolescent males with arginase deficiency demonstrating recurrent episodic hyperammonemia. Increased frequency of hyperammonemia has been reported in two other female arginase deficient patients [15,16]. Both patients had episodic hyperammonemia (values up to 222 μmol/L and 341 μmol/L, respectively) with vomiting, altered mental status, and coma consistently coinciding with their menstrual cycles. Both patients responded very well to hormonal treatment and/or hysterectomy. Patients could develop rapid hyperammonemia even after they are well into adulthood. Cederbaum et al. reported two male patients who died of hyperammonemia indicating multiple hospital admissions due to elevated ammonia in their teens and twenties [5]. Conclusion Arginase deficiency is the rarest yet least life-threatening urea cycle disorder. Although patients with arginase deficiency do not typically present with severe hyperammonemia, our patient has demonstrated more frequent episodes of hyperammonemia with increased severity since approximately ten years of age. Intravenous nitrogen scavenging medications were typically required. The long-term follow up of this patient may provide insights into the ideology why arginase deficiency may worsen as patients enter into adolescence and adulthood. In addition, our patient displayed a dramatic discrepancy between elevated ammonia and lack of matching clinical presentations. We were concerned that in some instances there might be a mismatch between the in vivo and ex vivo levels of ammonia. While the exact cause remains unknown, no obvious analytical concerns were found to explain the discrepancy. This is another example of the well-known but inexplicable and unpredictable difficulties with ammonia assessment in patients with urea cycle enzyme defects. We suspect that there may be other unknown factors that lead to this patient's recurrent hyperammonemia.
Due to various pre-analytical and analytical concerns about obtaining a reliable blood ammonia measurement, we strongly suggest that the chemistry laboratory works very closely with the clinical team when hyperammonemia is detected, especially for patients with metabolic diseases such as arginase deficiency, to avoid both treatment errors and irreversible brain damage. References [1] Gropman AL, Summar M, Leonard JV. Neurological implications of urea cycle disorders. J Inherit Metab Dis 2007;30:865-79. [2] Testai FD, Gorelick PB. Inherited metabolic disorders and stroke part 2: homocystinuria, organic acidurias, and urea cycle disorders. Arch Neurol 2010;67:148-53. [3] Brusilow SW, Horwich AL. Urea cycle enzymes. In: Valle, Beaudet, Vogelstein, Kinzler, Antonarakis, Ballabio, Childs, Sly, editors. The online metabolic and molecular bases of inherited disease. The McGraw Hill Companies; 2011. p. 98-9. [4] Crombez EA, Cederbaum SD. Hyperargininemia due to liver arginase deficiency. Mol Genet Metab 2005;84:243-51. [5] Grody WW, Kern RM, Klein D, Dodson AE, Wissman PB, Barsky SH, et al. Arginase deficiency manifesting delayed clinical sequelae and induction of a kidney arginase isozyme. Hum Genet 1993;91:1-5. [6] Vockley JG, Jenkinson CP, Shukla H, Kern RM, Grody WW, Cederbaum SD. Cloning and characterization of the human type II arginase gene. Genomics 1996;38: 118-23. [7] Morris Jr SM, Bhamidipati D, Kepka-Lenhart D. Human type ii arginase: sequence analysis and tissue-specific expression. Gene 1997;193:157-61. [8] Brusilow SW, Valle DL, Batshaw M. New pathways of nitrogen excretion in inborn errors of urea synthesis. Lancet 1979;2:452-4. [9] Enns GM, Berry SA, Berry GT, Rhead WJ, Brusilow SW, Hamosh A. Survival after treatment with phenylacetate and benzoate for urea-cycle disorders. N Engl J Med 2007;356:2282-92. [10] Maranda B, Cousineau J, Allard P, Lambert M. False positives in plasma ammonia measurement and their clinical impact in a pediatric population. Clin Biochem 2007;40:531-5. [11] da Fonseca-Wollheim F. Deamidation of glutamine by increased plasma gammaglutamyltransferase is a source of rapid ammonia formation in blood and plasma specimens. Clin Chem 1990;36:1479-82. [12] Serrano M, Ormazabal A, Vilaseca MA, Lambruschini N, Garcia-Romero R, Meavilla S, et al. Assessment of plasma ammonia and glutamine concentrations in urea cycle disorders. Clin Biochem 2011;44:742-4. [13] Consensus statement from a conference for the management of patients with urea cycle disorders. J Pediatr 2001;138:S1-5. [14] Barsotti RJ. Measurement of ammonia in blood. J Pediatr 2001;138:S11-9 [discussion S9-20]. [15] Grody WW, Chang RJ, Panagiotis NM, Matz D, Cederbaum SD. Menstrual cycle and gonadal steroid effects on symptomatic hyperammonaemia of urea-cycle-based and idiopathic aetiologies. J Inherit Metab Dis 1994;17:566-74. [16] Boles RG, Stone ML. A patient with arginase deficiency and episodic hyperammonemia successfully treated with menses cessation. Mol Genet Metab 2006;89:390-1.
Contents lists available at SciVerse ScienceDirect
Clinical Biochemistry journal homepage: www.elsevier.com/locate/clinbiochem
Recurrent unexplained hyperammonemia in an adolescent with arginase deficiency Yan Zhang a,⁎, Yuval E. Landau b, David T. Miller b, c, d, Deborah Marsden b, d, Gerard T. Berry b, d, Mark D. Kellogg c, e a
Department of Pathology and Laboratory Medicine, University of Rochester Medical Center, Rochester, NY, USA Division of Genetics, Children's Hospital Boston, Boston, MA, USA Department of Laboratory Medicine, Children's Hospital Boston, Boston, MA, USA d Department of Pediatrics, Harvard Medical School, Boston, MA, USA e Department of Pathology, Harvard Medical School, Boston, MA, USA b c
a r t i c l e
i n f o
Article history: Received 29 May 2012 Received in revised form 10 August 2012 Accepted 12 August 2012 Available online 23 August 2012 Keywords: Hyperammonemia Pre-analytical variations Arginase deficiency Urea cycle disorders
a b s t r a c t Objectives: This report investigates the etiology of recurrent episodic elevations in plasma ammonia in an adolescent male with arginase deficiency as there were concerns regarding pre-analytical and analytical perturbations of ammonia measurements. There were repeated discrepancies between the magnitude of his ammonia levels and the severity of his clinical signs of hyperammonemia. Patient and methods: The patient is a fourteen-year-old arginase-deficient male diagnosed at three years of age. Since 2008 (when he reached 10 years of age), there appeared to be an increase in the frequency of hospitalizations with elevated ammonia. A typical emergency visit with initial ammonia of 105 μmol/L (reference interval: 16–47 μmol/L) is illustrated. Pre-analytical and analytical procedures for the patient's sample handling were retrospectively examined. His ammonia levels were compiled since diagnosis. The frequency of his initial or peak ammonia levels greater than two times (94 μmol/L) or four times (188 μmol/L) the upper limit of normal was computed. Student t-test was used to calculate the significance of the differences before 2008 and since 2008. Results: One out of eleven and ten out of 19 hospitalizations had initial ammonia greater than two times normal before and after 2008, respectively. Both the patient’s overall ammonia and peak ammonia levels are significantly higher since 2008 (p value b 0.001 for both) than those before 2008. Conclusions: To our knowledge, few adolescent males with arginase deficiency experience recurrent episodes of hyperammonemia requiring intravenous nitrogen scavenging agents. We hope that this study provides new insights into the natural history of arginase deficiency and the management of such patients. © 2012 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Introduction The urea cycle consists of six consecutive enzymatic reactions to remove excess ammonia that is generated during the catabolism of nitrogen-containing compounds from the body [1]. Ammonia is in equilibrium with glutamine and is converted to urea through the urea cycle. Plasma ammonia concentrations are maintained within a fairly narrow range (reference interval [RI]: 16–47 μmol/L). Elevations are usually associated with acquired liver disease such as liver cirrhosis or severe hepatitis, or with certain organic acid disorders (propionic acidemia) or urea cycle disorders. Hyperammonemia, with one or several clinical signs of lethargy, confusion, irritability, vomiting, seizure, tremor, and coma, is a medical emergency requiring immediate medical
⁎ Corresponding author at: Dept. of Pathology and Laboratory Medicine, University of Rochester Medical Center, 601 Elmwood Avenue Box 608, Rochester, NY 14642, USA. Fax: +1 585 273 3003. E-mail address: [email protected] (Y. Zhang).
attention [1]. Prolonged exposure to high ammonia can lead to severe central nervous system damage and even death [1]. Arginase deficiency (also known as hyperargininemia, Mendelian Inheritance in Man number 207800), is caused by a deficiency of liver enzyme arginase I (E.C. 3.5.3.1). It is the least common urea cycle disorder with an estimated prevalence of 1 in 1,100,000 [2]. Arginase, the last enzyme in the urea cycle, catalyzes the conversion of arginine to urea and ornithine (Fig. 1). Urea is excreted in the urine, and ornithine is transported to the mitochondria to continue the cycle. Clinical manifestations of arginase deficiency are strikingly different from the other urea cycle disorders. As opposed to other urea cycle disorders that may present 24 to 72 h after birth or occasionally several days later into the extended neonatal period, arginase deficiency rarely presents in the neonatal period and first symptoms typically present in children between two and four years of age [3]. Hyperammonemia is not typically associated with arginase deficiency [4]. When hyperammonemic encephalopathy occurs, the plasma ammonia levels are three to four times normal value, with levels rarely greater than six times normal. These patients can become comatose
0009-9120/$ – see front matter © 2012 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.clinbiochem.2012.08.015
1584
Y. Zhang et al. / Clinical Biochemistry 45 (2012) 1583–1586
Fig. 1. Schematic of the urea cycle. Arginase is the last enzyme in this cycle which converts arginine to ornithine and urea.
when their ammonia levels rise to three to four times normal and may succumb to lethal brain edema [5]. Although not well studied, elevated arginase II, the mitochondria form of arginase expressed in kidney and brain, has been reported to be further induced by increased arginine levels in patients with arginase deficiency [6,7]. Arginase II may be the major contributor to the generally milder clinical presentations in these patients in comparison to other urea cycle disorders [5]. The less severe clinical course for arginase deficiency may also be due to the fact that arginase is at the last step of the urea cycle and two molecules of ammonium ions have already been accumulated into L-arginine by this point. In this report, we describe a male with arginase deficiency who had multiple episodes of hyperammonemia that required intravenous nitrogen scavenging medications between 10 and 14 years of age. In the majority of these episodes, he demonstrated less severe clinical presentation than expected based on the high ammonia levels. One typical emergency hospital visit is illustrated here during which he did not show typical clinical signs of hyperammonemia even at ammonia levels of 252 μmol/L and 242 μmol/L. This significant discrepancy triggered an investigation in the laboratory to identify 1) whether the ammonia level was falsely elevated due to pre-analytical and/or analytical problems, or 2) whether the discrepancy was related to the patient's underlying arginase deficiency.
However, his baseline is that of severe developmental disability, behavior problems and spastic diparesis. He has had 19 hospitalizations with elevated ammonia (>47 μmol/L) since 2008 with most of them requiring hydration and Ammonul (see description below) therapy. Eleven of these visits showed peak ammonia levels of more than four times normal (188 μmol/L). The average peak ammonia was 224 μmol/L (standard deviation [SD]=127 μmol/L). In the majority of these episodes, the clinical signs of hyperammonemia did not match the severity of his elevated plasma ammonia. In comparison, he had 11 hospital visits due to elevated ammonia between 2000 and 2008 and only three visits showed peak ammonia levels of two times normal (94 μmol/L) and none beyond four times normal (Fig. 2B). During this period, the average of his peak ammonia was 80 μmol/L with SD of 31 μmol/L. One of his typical emergency visits is described below. The patient presented to the emergency room with lethargy, emesis, and parental concerns about an “impending” metabolic crisis. The patient had complained of increased fatigue for several days, according to his mother. He had refused to eat by mouth on the day of admission and had been given hypercaloric and protein-free formula through his G-tube. Laboratory data obtained upon arrival for this visit indicated his ammonia was 105 μmol/L. Later analysis of the admission sample found arginine was 384 μmol/L (RI: 31–124 μmol/L) and decreased values for threonine, serine, isoleucine, and leucine. Glutamic acid and glutamine were within the reference intervals and gamma-glutamyltransferase (GTT) was 16 IU/L
Case description The patient is a fourteen-year-old male with arginase deficiency diagnosed at three years of age based on reduced arginase activity in red blood cells (November 2000). The patient had a history of severe intellectual disability, seizure disorder, asthma, gastrointestinal complications, and spastic diparesis. The patient was put on a protein-restricted diet (December 2000) soon after diagnosis. His most recent diet is composed of 0.4 g/kg of essential amino acids from medical food (formula), and 0.4 g/kg of intact protein from food. If he doesn't eat his full-recommended amount of protein, it is given as Lactaid milk through the gastrostomy tube (G-tube). The mother keeps detailed food records and typical foods include Cheese Nips, Cheez-Its, Wise onion rings, Light & Lively yogurt, Pringles, low protein cheese, and animal crackers. His height (by the age of 14) was 160.3 cm (29th percentile for age, z-score 0.55), weight is 63.6 kg (85th percentile for age, z-score 1.02), and body mass index is 24.8 kg/m2 (93rd percentile for age, z-score 1.44). Medications included sodium phenybutyrate (13.5 g/day), sodium benzoate (8.1 g/day), clonidine, speridone, Topamax, clinazepam, trazadine, Miralax, and Prilosec. The patient is generally well when there is no hyperammonemia.
Fig. 2. A. All ammonia measurements from the patient between 2000 and 2011. The patient was diagnosed with arginase deficiency in 2000 nearly at age of three. The p value indicated the difference of ammonia measurements before and after 2008. B. The initial and peak ammonia levels for patient's hospital visits due to elevated ammonia. The open squares are for peak ammonia and the open diamonds are for initial ammonia. The p value indicated the difference of peak ammonia before and after 2008.
Y. Zhang et al. / Clinical Biochemistry 45 (2012) 1583–1586
(RI: 12–55 IU/L). Hemolysis, icterus and lipemia indices for the sample were all below levels expected to cause interference. The patient was immediately evaluated in the emergency department and admitted to the medical intensive care unit for management of hyperammonemia. He was administered a continuous intravenous infusion of Ammonul (sodium benzoate/sodium phenylacetate; Medicis Pharmaceuticals Corp., Scottsdale, AZ), each at 5.5 g/m2/day in 10% glucose. An additional intravenous infusion of 20 mEq/L KCl was given so that the total fluid rate was 1.5 times maintenance. The sodium phenylbutyrate and sodium benzoate medication, given orally on a daily basis, were stopped. All other outpatient medications were continued. The components of Ammonul serve to enhance waste nitrogen excretion through the production of alternate waste nitrogen vehicles in the treatment of urea cycle disorder associated hyperammonemia. Specifically, phenylacetate and sodium benzoate are conjugated with glutamine and glycine, respectively, to form phenylacetylglutamine and hippuric acid which are then excreted through the kidneys [8,9]. Because the patient's plasma ammonia was only 105 μmol/L and the clinical signs of hyperammonemia were mild, it was decided to not administer the usually prescribed bolus dose of Ammonul (5.5 g/m 2/90 min). The therapy might also start at a lower dose based on the clinical situation. In the meantime, he was given ProPhree® (Abbott Nutrition, Columbus, OH), a protein-free formula, through the G-tube for additional calories. The protein-free diet was given for no more than 24–48 h after which he could start gradually getting his regular mix of essential amino acid formula with whole protein food. His plasma ammonia levels were closely monitored. The patient's general condition as well as his neurologic status appeared to improve shortly after admission and the lethargy did not worsen. The patient was not comatose, and he was not manifesting hyperventilation or seizure activity or any exacerbation in upper motor neuron signs such as the emergence of spontaneous ankle clonus. However, his ammonia level reached 252 μmol/L from a venous sample. An arterial blood sample obtained approximately 90 min after the venous sample measured an ammonia level of 242 μmol/L. These elevated ammonia levels did not reflect his improved clinical presentation. The following morning, the patient's lethargy had resolved, and the ammonia level had decreased to 65 μmol/L. It continued at about the same level for the next 60 h (Fig. 3).
Results and discussions Pre-analytical and analytical issues of ammonia measurement Ammonia is known to be one of the most difficult analytes to measure. Questions were raised for our patient since hyperammonemia is
Fig. 3. Plasma ammonia levels after patient's admission. Ammonia levels over a course of 60 h are shown. Dotted line, the reference interval for males, 16–47 μmol/L. Solid line, critical value in our hospital, 58 μmol/L.
1585
not typically associated with arginase deficiency patients and his clinical presentations did not match his elevated ammonia levels. The accuracy of ammonia measurements can be affected by either pre-analytical or analytical phases or both with the former being the major source of spuriously elevated ammonia results [10]. Delayed processing and hemolysis can lead to falsely elevated ammonia. Ammonia concentration may also be increased in vitro from the decomposition of glutamine seen in patients with high GTT activity [11], and elevated glutamine levels have also been reported in patients with urea cycle disorders, particularly patients with late-onset presentation [12]. The goal for reliable ammonia measurement is to minimize any release of ammonia from collected blood samples before analysis. The consensus statement from the conference on the management of patients with urea cycle disorders suggests that non-hemolyzed plasma samples should be collected in a pre-chilled, ammonia-free EDTA tube, kept on ice, and should be centrifuged within 15 min of collection [13]. Interpretation of ammonia results from patients with metabolic or liver pathology is significantly more challenging since samples from these patients can contain several sources of ammonia including ammonia formed in vitro at various rates and amounts. The analytical concerns associated with ammonia measurement by titration, colorimetric, fluorimetric, or gas-sensing electrode methodologies [14] have been largely addressed by the introduction of enzymatic assays based on glutamate dehydrogenase. In this report, the ammonia was measured on a Roche Cobas Integra800 instrument. The glutamate dehydrogenase specifically catalyzes the reduction of 2-oxoglutarate with NH4+ and NADPH to form L-glutamate and NADP. The concentration of NADPH consumed, which is determined by measuring the decrease in absorbance at 340 nm, is directly proportional to the ammonia concentration. For our patient, it was confirmed that all samples were collected in pre-chilled EDTA collection tubes and delivered on ice. Restrictions on tourniquet use, fist clenching and muscle activity were observed by staff performing blood collections. The samples were processed promptly, and the results were reported within 30 min. While the exact causes remain unknown, there were no obvious analytical or pre-analytical concerns for the elevated ammonia levels in this patient. The clinical features of arginase deficiency Arginase deficiency patients do not typically present with hyperammonemia in comparison to those with other urea cycle disorders. It's even rarer to see episodic hyperammonemia requiring intravenous Ammonul treatments in these patients. Our patient has had more severe ammonia elevations and increased frequency of hospital visits due to hyperammonemia since he entered the second decade of his life (2008) (Fig. 2A/B). The patient's ammonia levels have been significantly higher since 2008 (p valueb 0.001, N=89 before 2008, N=220 after 2008, Fig. 2A). The mean ammonia levels were 42 (SD=26) and 95 (SD=79) μmol/L for all the available ammonia measurements before and after 2008, respectively. Before 2008, there were no ammonia measurements beyond four times normal (maximum=131 μmol/L), while after 2008, 26 ammonia measurements have been recorded beyond that limit with six of them showing even more than 300 μmol/L (maximum=477 μmol/L) (Fig. 2A). Of his 19 hospital visits due to hyperammonemia since 2008, ten of them had initial ammonia more than two times normal and three visits had initial ammonia more than four times normal (Fig. 2B). Before 2008, however, only one hospital visit out of eleven showed an initial abnormal ammonia level of two times normal (Fig. 2B). The peak ammonia levels were also significantly higher after 2008 (mean = 224 μmol/L, SD = 127 μmol/L) than before 2008 (mean = 80 μmol/L, SD = 31 μmol/L, p b 0.001). Sixteen visits since 2008 showed peak ammonia levels more than two times normal and eleven visits more than four times normal while only three visits before
1586
Y. Zhang et al. / Clinical Biochemistry 45 (2012) 1583–1586
2008 had peak ammonia levels more than two times normal and none were four times normal or greater (Fig. 2B). His typical treatments during hyperammonemia involved intravenous ammonia reduction agents such as Ammonul as indicated in the Case description section. The patient was well when his ammonia was under control. Episodic hyperammonemia is well known for other types of urea cycle disorders. To our knowledge, there are few adolescent males with arginase deficiency demonstrating recurrent episodic hyperammonemia. Increased frequency of hyperammonemia has been reported in two other female arginase deficient patients [15,16]. Both patients had episodic hyperammonemia (values up to 222 μmol/L and 341 μmol/L, respectively) with vomiting, altered mental status, and coma consistently coinciding with their menstrual cycles. Both patients responded very well to hormonal treatment and/or hysterectomy. Patients could develop rapid hyperammonemia even after they are well into adulthood. Cederbaum et al. reported two male patients who died of hyperammonemia indicating multiple hospital admissions due to elevated ammonia in their teens and twenties [5]. Conclusion Arginase deficiency is the rarest yet least life-threatening urea cycle disorder. Although patients with arginase deficiency do not typically present with severe hyperammonemia, our patient has demonstrated more frequent episodes of hyperammonemia with increased severity since approximately ten years of age. Intravenous nitrogen scavenging medications were typically required. The long-term follow up of this patient may provide insights into the ideology why arginase deficiency may worsen as patients enter into adolescence and adulthood. In addition, our patient displayed a dramatic discrepancy between elevated ammonia and lack of matching clinical presentations. We were concerned that in some instances there might be a mismatch between the in vivo and ex vivo levels of ammonia. While the exact cause remains unknown, no obvious analytical concerns were found to explain the discrepancy. This is another example of the well-known but inexplicable and unpredictable difficulties with ammonia assessment in patients with urea cycle enzyme defects. We suspect that there may be other unknown factors that lead to this patient's recurrent hyperammonemia.
Due to various pre-analytical and analytical concerns about obtaining a reliable blood ammonia measurement, we strongly suggest that the chemistry laboratory works very closely with the clinical team when hyperammonemia is detected, especially for patients with metabolic diseases such as arginase deficiency, to avoid both treatment errors and irreversible brain damage. References [1] Gropman AL, Summar M, Leonard JV. Neurological implications of urea cycle disorders. J Inherit Metab Dis 2007;30:865-79. [2] Testai FD, Gorelick PB. Inherited metabolic disorders and stroke part 2: homocystinuria, organic acidurias, and urea cycle disorders. Arch Neurol 2010;67:148-53. [3] Brusilow SW, Horwich AL. Urea cycle enzymes. In: Valle, Beaudet, Vogelstein, Kinzler, Antonarakis, Ballabio, Childs, Sly, editors. The online metabolic and molecular bases of inherited disease. The McGraw Hill Companies; 2011. p. 98-9. [4] Crombez EA, Cederbaum SD. Hyperargininemia due to liver arginase deficiency. Mol Genet Metab 2005;84:243-51. [5] Grody WW, Kern RM, Klein D, Dodson AE, Wissman PB, Barsky SH, et al. Arginase deficiency manifesting delayed clinical sequelae and induction of a kidney arginase isozyme. Hum Genet 1993;91:1-5. [6] Vockley JG, Jenkinson CP, Shukla H, Kern RM, Grody WW, Cederbaum SD. Cloning and characterization of the human type II arginase gene. Genomics 1996;38: 118-23. [7] Morris Jr SM, Bhamidipati D, Kepka-Lenhart D. Human type ii arginase: sequence analysis and tissue-specific expression. Gene 1997;193:157-61. [8] Brusilow SW, Valle DL, Batshaw M. New pathways of nitrogen excretion in inborn errors of urea synthesis. Lancet 1979;2:452-4. [9] Enns GM, Berry SA, Berry GT, Rhead WJ, Brusilow SW, Hamosh A. Survival after treatment with phenylacetate and benzoate for urea-cycle disorders. N Engl J Med 2007;356:2282-92. [10] Maranda B, Cousineau J, Allard P, Lambert M. False positives in plasma ammonia measurement and their clinical impact in a pediatric population. Clin Biochem 2007;40:531-5. [11] da Fonseca-Wollheim F. Deamidation of glutamine by increased plasma gammaglutamyltransferase is a source of rapid ammonia formation in blood and plasma specimens. Clin Chem 1990;36:1479-82. [12] Serrano M, Ormazabal A, Vilaseca MA, Lambruschini N, Garcia-Romero R, Meavilla S, et al. Assessment of plasma ammonia and glutamine concentrations in urea cycle disorders. Clin Biochem 2011;44:742-4. [13] Consensus statement from a conference for the management of patients with urea cycle disorders. J Pediatr 2001;138:S1-5. [14] Barsotti RJ. Measurement of ammonia in blood. J Pediatr 2001;138:S11-9 [discussion S9-20]. [15] Grody WW, Chang RJ, Panagiotis NM, Matz D, Cederbaum SD. Menstrual cycle and gonadal steroid effects on symptomatic hyperammonaemia of urea-cycle-based and idiopathic aetiologies. J Inherit Metab Dis 1994;17:566-74. [16] Boles RG, Stone ML. A patient with arginase deficiency and episodic hyperammonemia successfully treated with menses cessation. Mol Genet Metab 2006;89:390-1.