`
`R E G U L A R A R T I C L E
`
`Diagnosis, symptoms, frequency and mortality of 260 patients with urea
`cycle disorders from a 21-year, multicentre study of acute
`hyperammonaemic episodes
`Marshall L Summar (Marshall.summar@vanderbilt.edu)1, Dries Dobbelaere2, Saul Brusilow3, Brendan Lee4
`1.Vanderbilt University Medical Center, Nashville, TN, USA
`2.Hospital University Medical Center of Lille, Lille Cedex, France
`3.Johns Hopkins Medical Center, Baltimore, MD, USA
`4.Howard Hughes Medical Institute, Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX, USA
`
`Keywords
`Drug treatment, Hyperammonaemia, Longitudinal
`study, Survival, Urea cycle disorders
`Correspondence
`Marshall L Summar, M.D., Department of Pediatrics
`and Molecular Physiology & Biophysics, Center for
`Human Genetic Research and Division of Medical
`Genetics, Vanderbilt University Medical Center,
`2215 Garland Ave., Suite 1175J Light Hall,
`Nashville, TN 37232-0007, USA.
`Tel: +615-322-6947 |
`Fax: +615-343-8619 |
`Email: Marshall.summar@vanderbilt.edu
`Received
`27 December 2007; revised 2 June 2008;
`accepted 25 June 2008.
`
`DOI:10.1111/j.1651-2227.2008.00952.x
`
`The authors do not have a financial conflict of
`interest in the publication of this material.
`
`Abstract
`Aim: A large longitudinal interventional study of patients with a urea cycle disorder (UCD) in
`hyperammonaemic crisis was undertaken to amass a significant body of data on their presenting
`symptoms and survival.
`Methods: Between 1982 and 2003, as part of the FDA approval process, data were collected on
`patients receiving an intravenous combination of nitrogen scavenging drugs (Ammonul R(cid:2)
`sodium
`phenylacetate and sodium benzoate (10%, 10%)) for the treatment of hyperammonaemic crises
`caused by urea cycle disorders.
`Results: A final diagnosis of a UCD was made for 260 patients, representing 975 episodes of
`hospitalization. Only 34% of these patients presented within the first 30 days of life and had a
`mortality rate of 32%. The most common presenting symptoms were neurological (80%), or
`gastrointestinal (33%). This cohort is the largest collection of patients reported for these diseases and
`the first large cohort in the United States.
`
`Conclusion: Surprisingly, the majority (66%) of patients with heritable causes of hyperammonaemia present
`beyond the neonatal period (>30 days). Patients with late-onset presenting disorders exhibited prolonged
`survival compared to the neonatal-presenting group.
`
`INTRODUCTION
`The urea cycle was first described in 1932 by Krebs and
`Henseleit (1). The urea cycle disorders (UCDs) result from
`defects in the clearance of excess nitrogen produced by
`the breakdown of protein and other nitrogen-containing
`molecules. The incidence of these disorders in the United
`States is roughly estimated to be at least 1/25 000 births but
`partial defects may mean that this number is much higher
`(2,3). In a Japanese study covering the period 1975–1995,
`the reported incidence was 1/46 000. Severe deficiency or
`total absence of activity of any of the first four enzymes
`in the urea cycle (carbamyl phosphate synthetase I, CPS-
`I; ornithine transcarbamylase, OTC; argininosuccinate syn-
`thetase, AS; argininosuccinate lyase, AL), or the cofactor
`producer N-acetyl glutamate synthase (NAGS), results in
`the accumulation of ammonia and other precursor metabo-
`lites during the first few days of life (4). In milder (or partial)
`urea cycle enzyme deficiencies, hyperammonaemia may be
`triggered by illness or stress at almost any time of life, result-
`ing in multiple mild elevations of plasma ammonia concen-
`tration (5,6). In these cases, the hyperammonaemias are less
`severe and the symptoms more subtle than in the patients
`with early-onset disease.
`The UCDs are considered a classic model for rare inborn
`errors of metabolism. Like many of these diseases, there are
`
`few patients at any particular treatment centre, and inci-
`dence and demographic data are sketchy at best. There has
`been a relative degree of uniformity in the treatment of acute
`episodes of hyperammonaemia in patients with UCDs dur-
`ing the past twenty years. This treatment involved the use of
`intravenous sodium phenylacetate and intravenous sodium
`benzoate as scavengers of excess nitrogen (2). The open In-
`vestigational New Drug (number 17 123) study during this
`time (1982–2003) required enrolment of patients receiving
`intravenous drug for the treatment of an acute episode of
`hyperammonaemia. The study was limited primarily to pa-
`tients with defects in the first three enzymes of the urea
`cycle (CPS-I, OTC, AS) with minimal inclusion of patients
`with arginase (ARG) deficiency or AL. As a part of the FDA
`approval process, a systematic chart review was done for
`these patients and data recorded on the presenting signs
`and symptoms, laboratory test parameters and any mor-
`talities related to these hyperammonaemic incidents. This
`publication reports on the most comprehensive dataset in
`existence on these patients. Using this information we have
`developed statistics on the age of onset of disease for these
`patients, presenting symptoms, disease incidence for the first
`three enzyme defects of the urea cycle and associated mor-
`tality rates. These data provide evidence-based insights into
`the nature of this particular group of rare disorders and
`
`1420
`
`Horizon Exhibit 2008
`Lupin v. Horizon
`IPR2018-00459
`
`1 of 7
`
`
`
`Acta Pædiatrica ISSN 0803–5253
`
`R E G U L A R A R T I C L E
`
`Diagnosis, symptoms, frequency and mortality of 260 patients with urea
`cycle disorders from a 21-year, multicentre study of acute
`hyperammonaemic episodes
`Marshall L Summar (Marshall.summar@vanderbilt.edu)1, Dries Dobbelaere2, Saul Brusilow3, Brendan Lee4
`1.Vanderbilt University Medical Center, Nashville, TN, USA
`2.Hospital University Medical Center of Lille, Lille Cedex, France
`3.Johns Hopkins Medical Center, Baltimore, MD, USA
`4.Howard Hughes Medical Institute, Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX, USA
`
`Keywords
`Drug treatment, Hyperammonaemia, Longitudinal
`study, Survival, Urea cycle disorders
`Correspondence
`Marshall L Summar, M.D., Department of Pediatrics
`and Molecular Physiology & Biophysics, Center for
`Human Genetic Research and Division of Medical
`Genetics, Vanderbilt University Medical Center,
`2215 Garland Ave., Suite 1175J Light Hall,
`Nashville, TN 37232-0007, USA.
`Tel: +615-322-6947 |
`Fax: +615-343-8619 |
`Email: Marshall.summar@vanderbilt.edu
`Received
`27 December 2007; revised 2 June 2008;
`accepted 25 June 2008.
`
`DOI:10.1111/j.1651-2227.2008.00952.x
`
`The authors do not have a financial conflict of
`interest in the publication of this material.
`
`Abstract
`Aim: A large longitudinal interventional study of patients with a urea cycle disorder (UCD) in
`hyperammonaemic crisis was undertaken to amass a significant body of data on their presenting
`symptoms and survival.
`Methods: Between 1982 and 2003, as part of the FDA approval process, data were collected on
`patients receiving an intravenous combination of nitrogen scavenging drugs (Ammonul R(cid:2)
`sodium
`phenylacetate and sodium benzoate (10%, 10%)) for the treatment of hyperammonaemic crises
`caused by urea cycle disorders.
`Results: A final diagnosis of a UCD was made for 260 patients, representing 975 episodes of
`hospitalization. Only 34% of these patients presented within the first 30 days of life and had a
`mortality rate of 32%. The most common presenting symptoms were neurological (80%), or
`gastrointestinal (33%). This cohort is the largest collection of patients reported for these diseases and
`the first large cohort in the United States.
`
`Conclusion: Surprisingly, the majority (66%) of patients with heritable causes of hyperammonaemia present
`beyond the neonatal period (>30 days). Patients with late-onset presenting disorders exhibited prolonged
`survival compared to the neonatal-presenting group.
`
`INTRODUCTION
`The urea cycle was first described in 1932 by Krebs and
`Henseleit (1). The urea cycle disorders (UCDs) result from
`defects in the clearance of excess nitrogen produced by
`the breakdown of protein and other nitrogen-containing
`molecules. The incidence of these disorders in the United
`States is roughly estimated to be at least 1/25 000 births but
`partial defects may mean that this number is much higher
`(2,3). In a Japanese study covering the period 1975–1995,
`the reported incidence was 1/46 000. Severe deficiency or
`total absence of activity of any of the first four enzymes
`in the urea cycle (carbamyl phosphate synthetase I, CPS-
`I; ornithine transcarbamylase, OTC; argininosuccinate syn-
`thetase, AS; argininosuccinate lyase, AL), or the cofactor
`producer N-acetyl glutamate synthase (NAGS), results in
`the accumulation of ammonia and other precursor metabo-
`lites during the first few days of life (4). In milder (or partial)
`urea cycle enzyme deficiencies, hyperammonaemia may be
`triggered by illness or stress at almost any time of life, result-
`ing in multiple mild elevations of plasma ammonia concen-
`tration (5,6). In these cases, the hyperammonaemias are less
`severe and the symptoms more subtle than in the patients
`with early-onset disease.
`The UCDs are considered a classic model for rare inborn
`errors of metabolism. Like many of these diseases, there are
`
`few patients at any particular treatment centre, and inci-
`dence and demographic data are sketchy at best. There has
`been a relative degree of uniformity in the treatment of acute
`episodes of hyperammonaemia in patients with UCDs dur-
`ing the past twenty years. This treatment involved the use of
`intravenous sodium phenylacetate and intravenous sodium
`benzoate as scavengers of excess nitrogen (2). The open In-
`vestigational New Drug (number 17 123) study during this
`time (1982–2003) required enrolment of patients receiving
`intravenous drug for the treatment of an acute episode of
`hyperammonaemia. The study was limited primarily to pa-
`tients with defects in the first three enzymes of the urea
`cycle (CPS-I, OTC, AS) with minimal inclusion of patients
`with arginase (ARG) deficiency or AL. As a part of the FDA
`approval process, a systematic chart review was done for
`these patients and data recorded on the presenting signs
`and symptoms, laboratory test parameters and any mor-
`talities related to these hyperammonaemic incidents. This
`publication reports on the most comprehensive dataset in
`existence on these patients. Using this information we have
`developed statistics on the age of onset of disease for these
`patients, presenting symptoms, disease incidence for the first
`three enzyme defects of the urea cycle and associated mor-
`tality rates. These data provide evidence-based insights into
`the nature of this particular group of rare disorders and
`
`1420
`
`C(cid:2)2008 The Author(s)/Journal Compilation C(cid:2)2008 Foundation Acta Pædiatrica/Acta Pædiatrica 2008 97, pp. 1420–1425
`
`2 of 7
`
`
`
`Summar et al.
`
`Twenty-one-year study of UCDs and hyperammonaemia
`
`challenge some of the assumptions typically made about
`these patients.
`
`SUBJECTS AND METHODS
`The data source for this study was an open-label, uncon-
`trolled study in patients with hyperammonaemia due to
`UCDs. Data collected from February 10, 1982 through May
`31, 2003 were included in the analyses for this article;
`the study was closed on March 31, 2005 after marketing
`approval for sodium phenylacetate and sodium benzoate
`(NaPA/NaBZ) injection 10%/10% (Ammonul R(cid:2), Ucyclyd
`Pharma, Scottsdale, AZ, USA) was granted by the Food and
`Drug Administration (FDA). Patients were originally treated
`under an investigational new drug application (IND number
`17 123) sponsored, prior to 1999, by Dr. Saul W. Brusilow
`of the Johns Hopkins School of Medicine; the IND was
`transferred to Ucyclyd Pharma in 1999. Written informed
`consent, as appropriate for each institution, was obtained
`from each patient or their legal authorized representative.
`Institutional Review Board (IRB) approval was obtained as
`appropriate at that time at the treating institutions. To the
`best knowledge of the authors and other experts in the field,
`this study was the only significant source of drug available
`to treat these patients during the period of 1982 to 2003.
`The original protocol called for the treatment of patients
`with CPS-I, OTC or AS deficiency, and not AL or ARG
`deficiency, and therefore the numbers for AL and ARG de-
`ficiency in this report do not reflect their true incidences in
`the patient population.
`An episode was defined as a hospitalization for hyperam-
`monaemia during which the patient received NaPA/NaBZ
`Injection 10%/10% as rescue treatment. Data collected as
`part of the FDA study included plasma ammonia levels and
`presenting signs and symptoms. Episodes described herein
`occurred at 115 hospitals in the United States and Canada.
`Data were collected on patients only during their time in
`hospital and no data were collected on these patients in the
`intervals when they were discharged from the hospital.
`
`A specific urea cycle diagnosis was made for each patient
`by the treating physician. Acceptable diagnostic methods
`used included liver biopsy with enzymatic analysis (all dis-
`orders), mutation analysis (all disorders), allopurinol chal-
`lenge test combined with hyperammonaemia and plasma
`amino acid analysis (OTC and CPS-I), plasma amino acid
`analysis (AS and AL) and erythrocyte assay (ARG).
`
`Statistical methods
`The data presentation for this article consists of descrip-
`tive statistics including means, standard deviations and me-
`dians for continuous variables and frequency counts and
`percentages for categorical variables. Variation is reported
`as standard deviation. Missing data are excluded from the
`summaries and percentage calculations. Kaplan–Meier sur-
`vival plots are also provided. Patients were censored at the
`hospital discharge date from the last episode experienced to
`avoid missing data errors. All data analyses were performed
`using SAS R(cid:2) (Cary, NC, USA) version 8.2.
`
`RESULTS
`Three hundred and sixteen patients were treated for 1045
`episodes of hospitalization between 1982 and 2003; how-
`ever, 56 patients did not have definitive final diagnoses of
`a specific UCD and were thus not included in the analyses
`for this article. A final diagnosis of a UCD was made for 260
`patients, representing 975 episodes of hospitalization.
`
`Patient diagnoses and age at presentation
`The most frequent diagnosis was OTC deficiency in 142
`of 260 patients (55%), followed by AS deficiency (citrul-
`linaemia I) in 70 patients (27%), CPS-I deficiency in 36
`patients (14%), AL deficiency in 7 patients (3%) and ARG
`deficiency in 2 patients (<1%).
`Figure 1 shows the age of each patient at the first hos-
`pitalization for hyperammonaemia, by each UCD diagno-
`sis and by all diagnoses. Overall, the median age at first
`
`Figure 1 Number of patients by diagnosis and age at first episode. Each patient is counted once at the age the first episode of hyperammonaemia was reported.
`
`C(cid:2)2008 The Author(s)/Journal Compilation C(cid:2)2008 Foundation Acta Pædiatrica/Acta Pædiatrica 2008 97, pp. 1420–1425
`
`1421
`
`3 of 7
`
`
`
`Twenty-one-year study of UCDs and hyperammonaemia
`
`Summar et al.
`
`presentation was 2.0 years and ranged from 1 day to
`53 years. Of the total of 260 patients diagnosed with a UCD,
`34% were aged 0 to 30 days, 18% were aged 31 days to
`2 years, 28% were aged >2 to 12 years, 10% were aged >12
`to 16 years, and 10% were older than 16 years.
`The typical age at first presentation varied among the
`different diagnoses. Males with OTC deficiency tended to
`present earlier (median of approximately 6 months) than
`females (median age of 10 years). Of the 69 male patients
`with OTC deficiency, 46% presented at age 0 to 30 days,
`19% at 31 days to 2 years, 22% at >2 to 12 years, and 13% at
`>12 years. As expected for this X-linked disorder, onset dur-
`ing infancy was the typical pattern; however, more than half
`of the males had late-onset OTC deficiency. Among the 73
`female patients with OTC deficiency, onset during child-
`hood or adolescence was the typical pattern, although only
`4% presented at age 0 to 30 days, compared to 16% between
`the age of 31 days to 2 years, 51% at >2 to 12 years, 10% at
`>12 to 16 years, and 19% at >16 years.
`For patients with CPS-I and AS deficiencies, there was
`no typical age at first presentation. For both of these disor-
`ders, the mean age of first presentation was during young
`childhood (median approximately 11 months) for CPS-I de-
`ficiency, and a median of 2.0 years for AS deficiency. Sub-
`stantial percentages of patients had their first episodes at
`age 0 to 30 days (27% for CPS-I deficiency and 37% for AS
`deficiency), at 31 days to 2 years (30% for CPS-I and 14%
`for AS deficiency), and at >2 to 12 years (19% for CPS-I
`and 33% for AS deficiency).
`For AL deficiency, five of seven patients presented at age
`0 to 30 days, one patient at 6 years, and one patient at
`14 years, and for ARG deficiency, one patient presented at
`3 days and the other at 19 years.
`
`Recurrence of hyperammonaemia
`Figure 2 shows the number of hyperammonaemic episodes
`by diagnosis and by age. Many patients had multiple
`episodes of hyperammonaemia, and thus, overall, the total
`
`number of hyperammonaemic episodes is far greater than
`the number of patients. For patients with AS or CPS-I de-
`ficiencies and males with OTC deficiency, approximately
`70% of episodes occurred between the age of 31 days and
`12 years. For females with OTC deficiency, 92% of episodes
`occurred at age >2 years.
`Among patients with five or more reported episodes, the
`average number of episodes per year was 2.2 for patients
`with AS deficiency, 2.9 for males with OTC deficiency, 2.4
`for females with OTC deficiency and 2.8 for patients with
`CPS-I deficiency. The number of episodes per year varied
`considerably among individual patients, ranging from 0.6
`episodes per year in a patient with AS deficiency (nine
`episodes from the age of 3 to 17 years) to eight episodes
`per year in a female with OTC deficiency (32 episodes from
`the age of 18 to 21). One patient with CPS-I deficiency had
`eight hyperammonaemic episodes in the first year of life. The
`greatest number of episodes reported in any single patient
`was 77 in a female with OTC deficiency. She had an average
`of 3.5 episodes per year between the age of 2 and 23.
`
`Signs and symptoms of a hyperammonaemic episode in
`patients with UCDs
`Most hyperammonaemic episodes (58% of all episodes)
`were preceded by reported illness. Other known events pre-
`ceding episodes included non-compliance with diet (15%),
`non-compliance with medication (10%), and major life
`events (such as surgery, accidents, school stress, pregnancy,
`etc.) (10%).
`Presenting signs and symptoms of a hyperammonaemic
`episode are summarized in Table 1. In the majority of
`episodes (76%), patients already had neurological symp-
`toms at the time of hospital admission. A decreased level of
`consciousness (symptoms included fatigue, lethargy, drowsi-
`ness, unresponsiveness, coma and obtundation), and/or ab-
`normal motor function (symptoms included slurred speech,
`tremors, weakness, decreased or increased muscle tone and
`ataxia), were more frequent at the first episode than at
`
`Figure 2 Number of episodes by diagnosis and age at first episode. Each patient with multiple episodes is counted multiple times at each age during which an
`episode of hyperammonaemia was reported.
`
`1422
`
`C(cid:2)2008 The Author(s)/Journal Compilation C(cid:2)2008 Foundation Acta Pædiatrica/Acta Pædiatrica 2008 97, pp. 1420–1425
`
`4 of 7
`
`
`
`Summar et al.
`
`Twenty-one-year study of UCDs and hyperammonaemia
`
`Table 1 Frequency of presenting signs and symptoms
`
`First episode
`n = 260
`
`No. of episodes
`
`Subsequent
`episode
`n = 715
`
`Any episode
`n = 975
`
`240 (92.3%)
`
`670 (93.7%)
`
`910 (93.3%)
`
`208 (80.0%)
`164 (63.1%)
`
`532 (74.4%)
`365 (51.0%)
`
`740 (75.9%)
`529 (54.3%)
`
`83 (31.9%)
`78 (30.0%)
`25 (9.6%)
`17 (6.5%)
`
`85 (32.7%)
`50 (19.2%)
`24 (9.2%)
`9 (3.5%)
`9 (3.5%)
`0
`13 (5.0%)
`
`118 (45.4%)
`75 (30.0%)
`16 (6.2%)
`17 (6.5%)
`10 (3.8%)
`7 (2.7%)
`3 (1.2%)
`15 (5.8%)
`
`35 (13.5%)
`17 (3.5%)
`11 (4.2%)
`5 (1.9%)
`2 (0.8%)
`0
`2 (0.8%)
`
`268 (37.5%)
`155 (21.7%)
`34 (4.8%)
`65 (9.1%)
`
`340 (47.6%)
`242 (33.8%)
`75 (10.5%)
`58 (8.1%)
`31 (4.3%)
`9 (1.3%)
`73 (10.2%)
`
`299 (41.8%)
`252 (35.2%)
`57 (8.0%)
`12 (1.7%)
`3 (0.4%)
`6 (0.8%)
`1 (0.1%)
`27 (3.8%)
`
`84 (11.7%)
`56 (7.8%)
`26 (3.6%)
`1 (0.1%)
`2 (0.3%)
`1 (0.1%)
`3 (0.4%)
`
`351 (36.0%)
`233 (23.9%)
`59 (6.1%)
`82 (8.4%)
`
`425 (43.6%)
`292 (29.9%)
`99 (10.2%)
`67 (6.9%)
`40 (4.1%)
`9 (0.9%)
`86 (8.8%)
`
`417 (42.8%)
`330 (33.8%)
`73 (7.5%)
`29 (3.0%)
`13 (1.3%)
`13 (1.3%)
`4 (0.4%)
`42 (4.3%)
`
`119 (12.2%)
`73 (7.5%)
`37 (3.8%)
`6 (0.6%)
`4 (0.4%)
`1 (0.1%)
`5 (0.5%)
`
`0
`
`18 (2.5%)
`
`18 (1.8%)
`
`Symptom
`
`Any
`
`Neurological
`Decreased level of
`consciousness
`Altered mental status
`Abnormal motor function
`Seizures
`∗
`Other
`
`Gastrointestinal
`Vomiting
`Poor feeding
`Diarrhoea
`Nausea
`Constipation
`†
`Other
`
`Co-morbid conditions
`Infection
`Respiratory
`Renal
`Cardiovascular
`Haematological
`Hepatic
`‡
`Other
`
`General/constitutional
`Fever
`Dehydration
`Hypothermia
`Weight loss
`Failure to thrive
`§
`Other
`
`Routine check-up
`∗
`
`Includes headache, stroke, cerebral oedema, cerebral haemorrhage, cortical
`blindness, hemianopsia with scotoma and burning eyes.
`†
`Includes abdominal pain, gastritis, duodenitis,
`ileus, gastro-oesophageal
`reflux, gastrointestinal (GI) dysmotility,
`intestinal obstruction,
`irritable bowel
`syndrome, pancreatitis, GI bleed, duodenal atresia, gastroparesis, pyloric ulcer
`and pyloric stenosis.
`‡
`Includes rash, fracture, burn, surgery, cholelithiasis, ovarian cyst, hernia and
`necrosis of femoral heads.
`§
`Includes hypovolaemia, paleness, puffiness and oedema.
`
`subsequent episodes. Altered mental status (symptoms in-
`cluded irritability, tantrums, strange behaviour or speech,
`dizziness, confusion, agitation and combativeness) were
`slightly more frequent at subsequent episodes than at the
`first episode.
`Other notable presenting signs and symptoms included
`infection (reported in 34% of episodes) and vomiting (30%).
`Both sets of signs and symptoms were somewhat more fre-
`quent at subsequent episodes than at the first episode, par-
`ticularly vomiting.
`
`It should also be noted that in some patients (18 hyper-
`ammonaemic episodes), an elevated plasma ammonia level,
`which received treatment, was noted during a routine office
`visit and was not accompanied by symptoms typical of hy-
`perammonaemia. In 9 of the 18 episodes for which hyper-
`ammonaemia was discovered during a routine office visit,
`no other symptoms were reported, while in the other nine
`episodes, mild neurological symptoms, such as irritability,
`drowsiness or restlessness, were noted during neurological
`examination but had not been reported previously. It is pos-
`sible that in these circumstances, the high levels of ammonia
`may reflect an artefact of specimen handling.
`
`Survival of patients with UCDs
`Figure 3 shows a Kaplan–Meier graph of survival by age at
`the first episode of hyperammonaemia. Survival time is cal-
`culated as the amount of time between the discharge date
`of the last episode and the admission date from the first
`episode. Patients who presented with the first hyperammon-
`aemic episode at age between 0 and 30 days had the worst
`outcome—only 35% of patients were still alive at the final
`follow-up time point (approximately 11 years after the start
`of the study period). Of the 26 deaths that occurred in this
`group, 22 were within the first 2 months of life and only one
`patient after 3 years at the age of 6 years. Patients presenting
`in late infancy had the best outcome, with 87% remaining
`alive at the final follow-up time point. Only two patients in
`this group died, one at approximately 10 months of age and
`the other at approximately 2.5 years after the first episode.
`Of the patients presenting at >2 to 16 years of age, 78%
`were alive at the final follow-up time point. Ten of the 12
`deaths in this group occurred within the first 50 days of ini-
`tial presentation, one approximately 2 years later, and one
`approximately 7 years later.
`Figure 4 shows a Kaplan–Meier graph of survival by di-
`agnosis (data are not presented for AL and ARG) . Percent
`survival at the final follow-up time point was 78% among
`patients with AS deficiency, 74% among females with OTC
`deficiency, 61% among patients with CPS-I deficiency, and
`53% among males with OTC deficiency. For all diagnoses,
`most deaths occurred within the first 2 months of initial pre-
`sentation (16 of 20 deaths for males with OTC deficiency,
`eight of nine deaths for females with OTC deficiency, four
`of seven deaths for patients with CPS-I deficiency, and five
`of six deaths for patients with AS deficiency.
`
`DISCUSSION
`This report is the second large overview of data collected
`from the United States on the incidence of UCDs com-
`bined with clinical presentation and outcome data, and is
`the largest report on UCDs to date. These data also re-
`flect the survival rates using the current treatment methods
`available for hyperammonaemic crisis (nitrogen scavenging
`compounds). These data are surprising in the fact that they
`contradict the commonly held belief that UCDs are pri-
`marily disorders of the newborn period. As our data show,
`even when female OTC carriers are excluded, the most
`
`C(cid:2)2008 The Author(s)/Journal Compilation C(cid:2)2008 Foundation Acta Pædiatrica/Acta Pædiatrica 2008 97, pp. 1420–1425
`
`1423
`
`5 of 7
`
`
`
`Twenty-one-year study of UCDs and hyperammonaemia
`
`Summar et al.
`
`Figure 3 Kaplan–Meier survival by age at first episode.
`
`Figure 4 Kaplan–Meier survival by diagnosis.
`
`common first presentation period for all diseases is outside
`the newborn period. The age distributions seen in these data
`contrast with the traditional view of UCDs as being primarily
`diseases of the newborn. Although 0 to 30 days was the most
`common age at first presentation, two-thirds of all patients
`had late-onset disease, and childhood (age 2 to 12 years)
`appears to be another vulnerable time period for the first
`hyperammonaemic episode. Accordingly, physicians caring
`for adults and older children should strongly consider a di-
`agnosis of a UCD when confronted with the appropriate
`symptoms.
`The data presented in this publication represent the orig-
`inal 316 patients and 1045 episodes of hyperammonaemia
`that were analyzed for the U.S. New Drug Application and
`that are described in the current Ammonul package insert.
`The data presented in a recent New England Journal of
`Medicine paper by Enns et al. (7) used a slightly larger
`dataset of 389 patients and 1294 episodes of hospitaliza-
`tion that were included in the safety update provided to the
`FDA at the time of the approval of the product. Analyses
`
`for both articles excluded patients without a diagnosis of
`a UCD. This manuscript uses this dataset to focus on the
`clinical presentations, morbidity and mortality rather than
`the acute response to drug. Slight differences between the
`two articles in survival are found because of the additional
`patients collected after the cut-off for this work.
`It should also be noted that the highest mortality peaks
`for all disorders occur very close to the initial presentation.
`This suggests that there is a subset of patients that have such
`severe disease that they are resistant to therapy. Alterna-
`tively, delayed initial diagnosis and the consequent magni-
`tude of hyperammonaemia may adversely affect the outcome
`of the initial hospitalization for hyperammonaemia. Of the
`43 deaths in these 260 patients, 27 occurred after the with-
`drawal of life support.
`There was no strict definition of an elevated plasma am-
`monia level used in this study. It was the responsibility of the
`individual investigator to decide when treatment should be
`initiated. The investigator was asked only to check a box on
`the case report form noting whether the hyperammonaemic
`
`1424
`
`C(cid:2)2008 The Author(s)/Journal Compilation C(cid:2)2008 Foundation Acta Pædiatrica/Acta Pædiatrica 2008 97, pp. 1420–1425
`
`6 of 7
`
`
`
`Summar et al.
`
`Twenty-one-year study of UCDs and hyperammonaemia
`
`episode was preceded by illness, non-compliance with diet
`or medication, or major life event. A comments field allowed
`an investigator to report if the elevated plasma ammonia
`level was observed at a routine office visit and/or that the
`patient was asymptomatic.
`The incidence data for the disorders targeted in this study
`group are very similar to those seen in the Japanese popula-
`tion in a study performed by Uchino et al. (3) on 216 cases.
`The biggest discrepancies were fewer males with OTC defi-
`ciency in the U.S. data (26% vs. 43%) and more patients with
`AS deficiency (27% vs. 11%). This may reflect detection and
`reporting rates of different disorders although similar labo-
`ratory methods are used. There has not been a report of a
`common mutation for either disease in either population at
`this time so we cannot account for it in this regard. Since
`the majority of these studies were performed before the ad-
`vent of expanded newborn screening, there should not be a
`detection difference based purely on methodology.
`The survival data from this study are more encouraging
`than the commonly held opinion might be for these disor-
`ders. In all disease groups, at least 50% of the patients were
`alive at 5 years. Mortality seems to correlate closely with age
`at presentation, with the newborn group having the lowest
`survival statistics. These data also justify the consideration
`of alternative therapies such as orthotopic liver transplanta-
`tion in those enzyme deficiencies such as OTC and CPS-I
`that present in the neonatal period.
`For the clinical presentation of these patients it was ex-
`pected and confirmed that neurological signs would be the
`most common presenting feature. It was surprising that only
`63% of patients had a reported decreased level of conscious-
`ness at the first episode and only 19% presented with vomit-
`ing. A decreased level of consciousness (symptoms included
`fatigue, lethargy, drowsiness, unresponsiveness, coma and
`obtundation), and/or abnormal motor function (symptoms
`included slurred speech, tremors, weakness, decreased or in-
`creased muscle tone and ataxia), were more frequent at the
`first episode than at subsequent episodes. This may be be-
`cause patients were younger at the first episode, or because
`treatment was sought sooner at subsequent episodes (after
`a definitive diagnosis had been established), before more
`severe symptoms such as coma and ataxia had developed.
`Altered mental status (symptoms included irritability,
`tantrums, strange behaviour or speech, dizziness, confusion,
`agitation and combativeness) were slightly more frequent at
`subsequent episodes than at the first episode. Such symp-
`toms may be more frequent among older patients, or may
`have been reported less frequently at the first episode be-
`cause they were overshadowed by more serious symptoms
`such as coma and ataxia.
`The high incidence of infection (34%) as a co-morbid con-
`dition underscores the importance of careful ‘sick-day’ mon-
`
`itoring and treatment of these patients. Moreover, these data
`support the long-time anecdotal observation that catabolic
`stress, e.g. from a viral illness, is a more significant risk factor
`for hyperammonaemia than an increased dietary intake of
`nitrogen.
`These data may not contain all of the presenting symptoms
`of these patients, but a careful review of each medical record
`was made to extract the data.
`The low frequency of patients with AL deficiency (argini-
`nosuccinic aciduria) differs from other reports; however, as
`described in the methods, these patients were not included
`in the original study design for treatment.
`In summary, UCDs represent a model for studies of
`the incidence and prevalence of rare inborn errors of
`metabolism. The NIH has sponsored a long-term in-depth
`longitudinal study of these patients to answer many of the
`questions about neurological development, survival, qual-
`ity of life and growth that are not answered in this study.
`More information about this study and an enrolment site
`for patients can be found at: http://rarediseasesnetwork.
`epi.usf.edu/ucdc/index.htm.
`
`ACKNOWLEDGEMENTS
`B.L. and M.S. are supported by the Rare Disease Clinical
`Research Center Urea Cycle Grant (RR19453-01), B.L. is
`supported by R01 DK54450 and Baylor College of Medicine
`Mental Retardation and Developmental Disabilities Center
`(MRDDC). We would like to thank David Varnam for the
`excellent editing work in this manuscript. We would like to
`thank Erin Glynn and Joe Mauney for their help in database
`and statistical work.
`
`References
`
`1. Krebs HA, Henseleit K. Untersuchungen uber die
`harnstoffbildung im tierkorper. Hoppe-Seyler’s Z Physiol
`Chem 1932; 210: 325–32.
`2. Brusilow SW, Maestri