Long-term Follow-up of Neonatal Mitochondrial Cytopathies: A Study of 57 Patients

Long-term Follow-up of Neonatal Mitochondrial Cytopathies:

A Study of 57 Patients

A. Garcı´a-Cazorla, PhD*; P. De Lonlay, PhD*; M.C. Nassogne, PhD*; P. Rustin, PhD‡; G. Touati, PhD*; and J.M. Saudubray, PhD*

ABSTRACT. Objectives. We sought to determine the long-term clinical and biochemical outcome of newborns with mitochondrial cytopathies (MCs) and to identify possible prognostic factors that may modify the course of these diseases.

Material and Methods. Fifty-seven newborns with MCs were identified in a retrospective review (1983– 2002). We defined 2 different outcome categories: clinical (neurologic, hepatic, myopathic, and multiorganic) and biochemical (lactate level normalization or initially nor-mal remaining unchanged, decreased but not nornor-mal- normal-ized, and persistently high). We used 2 different statisti-cal approaches: (1) survival studies depending on the initial symptoms and lactate and enzymatic deficiencies using the Kaplan-Meier method; and (2) the same variables compared with different survival age groups and clinical and bio-chemical outcome categories using the␹2_test.

Results. Thirty-three patients died (57.8%), 12 remain alive (21%), and 12 were lost in the follow-up; 6 of them are currently older than 4 years. Most of the patients manifested multiorganic disease (64.8%) and high lactate level (77.1%) over time. Children surviving to 2.5 to 3 years of age were more likely to survive for a long period of time. Initial neurologic and hepatic presentation in-creased the risk to develop neurologic disease and severe persistent hyperlactacidemia, respectively. Initial severe hyperlactacidemia and combined enzyme deficiencies were significant risk factors for higher mortality and multiorganic disorders. Two patients with exclusively myopathic outcome are alive and cognitively normal at 12 years of life.

Conclusions. Children with neonatal-onset MCs have very high mortality and poor prospects. However, some with life-threatening presentations may gradually im-prove, giving rise to less severe diseases. Those with exclusively myopathic symptoms have a better prognosis.

Pediatrics 2005;116:1170–1177; long-term follow up, mito-chondrial disorders, neonatal morbidity, neonatal mortality, hyperlactacidemia.

ABBREVIATIONS. C, electron transport chain complex; CPK, cre-atine phosphokinase; MC, mitochondrial cytopathy.

M

great variety of symptomatology and, in most of the cases, rapidly progressive course are well known because of the major involvement of high-rate aero-bic metabolism in most organs. Furthermore, neona-tal presentation is quiet common because of the high energy requirements of the growing newborn. Clin-ical unspecific features such as hypotonia, feeding difficulties, failure to thrive, and seizures are very frequent early-onset presentation symptoms.4

Biochemically, markers such as high levels of plas-matic lactate, pyruvate, and some amino acids such as alanine, proline, and glutamine, as well as raised urinary lactate and other Krebs-cycle organic acid levels, are also well-recognized impaired redox-func-tion features.5

By contrast, relatively little is known about the long-term clinical and biological outcome. A rapid and fatal course is normally suspected in neonatal presentations of these diseases. Nevertheless, no long-term studies have been reported in the litera-ture.

Here we report the clinical and biochemical out-come over a long period of time of a group of pa-tients diagnosed and managed in our hospital. We examined the long-term outcome and its relation with the clinical presentation, with enzymatic defi-ciency and initial plasmatic lactate level as possible contributing prognosis factors.

PATIENTS AND METHODS Patients’ Characteristics

Fifty-seven patients (30 males and 27 females) with respiratory chain disorders of neonatal onset (age range: 0 –30 days; mean: 6.35 days; SD: 8.5 days) were diagnosed at the metabolic unit of the Hoˆpital Necker Enfants Malades from 1983 to 2002. A sugges-tive clinical picture associated with high plasmatic lactate level in the majority of cases (50 of 57 [87.7%]; mean: 9.7 mmol/L), abnor-mal plasma amino acids (44 of 57 [77.1%]) consisting of increased alanine (46 of 57 [80.7%]), proline (23 of 57 [40.3%]), and glutamine (17 of 57 [29.8%]), and abnormal urine organic acids (43 of 57 [75.4%]; consisting of increased lactate level in 43 of 57 [75.4%] and other Krebs-cycle metabolites in 40 of 57 [70.1%]) led us to suspect the diagnosis.

Concerning the clinical symptoms, we mostly found multior-ganic presentations. Hypotonia, tachypnea, and feeding difficul-ties formed the most common pattern of association, followed by neurologic and hepatodigestive symptoms.

Depending on these different presentation types, we estab-lished the following clinical categories: multiorganic (53 of 57

From the *Metabolic Diseases Unit, Department of Pediatrics, Centre Hos-pitalier Universitaire Necker Enfants-Malades, Paris, France; and ‡Institut National de la Sante´ et de la Recherche Me´dicale, Research Units U375 and 30, Paris, France.

Accepted for publication Jan 21, 2005. doi:10.1542/peds.2004-2407 No conflict of interest declared.

Address correspondence to A. Garcı´a-Cazorla, PhD, Neurology Service, Hospital Sant Joan de De´u, Passeig Sant Joan de De´u, 2, 08950 Esplugues, Barcelona, Spain. E-mail: agarcia@hsjdbcn.org

[92.9%]); association of hypotonia, tachypnea, and feeding diffi-culties (25 of 57 [43.8%]); neurologic (21 of 57 [36.8%]); hepatodi-gestive (15 of 57 [26.3%]); failure to thrive (8 of 57 [14%]); and cardiac (3 of 57 [5.2%]). Most of the patients presenting with cardiac disorders are evaluated and followed in the cardiology unit; therefore, there is a very low proportion of them in our series.

Methods

Lactate level was measured in 1- to 1.5-mL blood samples obtained by venipuncture, avoiding hand exercise before the sam-pling, and collected in heparinized tubes. From these tubes, 0.5 mL of whole blood that was accurately pipetted was immediately deproteinized and transferred into 1-mL ice-cold perchloric acid (1 mmol/L) tubes. After thorough mixing, samples could be stored at⫺20°C and thawed immediately before analysis by enzymatic semiautomated methods.

Initial blood lactate level ranged from 1.7 to 25.5 mmol/L (mean: 9.7 mmol/L; SD: 6.65 mmol/L). The following lactate-level categories were defined: normal lactate level (⬍2.4 mmol/L), 7 patients (12.2%); mild hyperlactacidemia (2.5–2.9 mmol/L), 1 tient (1.7%); moderate hyperlactacidemia (3–5.9 mmol/L), 11 pa-tients (19.2%); and severe hyperlactacidemia (⬎6 mmol/L), 38 cases (66.6%).

The diagnosis was confirmed by measurements of respiratory chain enzymes in muscle, fibroblasts, lymphocytes, or liver by spectroscopy and polarography. Twenty-four electron transport chain complex (C)I, 12 CI⫹CIV, 7 CIV, 7 generalized (6 of which were associated with DNA mitochondrial depletion), 3 CII, 1 CIII, 1 CV, 1 CIII⫹CIV deficiencies, and 1 patient diagnosed on muscle morphology criteria, were observed. We defined 5 enzymatic cat-egories: CI; CI⫹CIV; CIV; generalized; and others (CII, CIII, CV, CIII⫹IV, muscle morphology).

Other than mitochondrial DNA depletions, genetic studies only found a mutation in the ATP synthase gene (p8993) in 1 patient. Mitochondrial DNA depletions (depletion range in our patients: 20 –75%) were detected by using Southern blot analysis in muscle (4 patients who presented with neurologic and hepatic

symp-toms), liver (1 patient with liver failure), and heart (1 patient with dilated cardiomyopathy).

Treatment and Clinical Follow-up

All the patients were initially treated with different cofactors (thiamin, riboflavin, carnitine), most of them with bicarbonate (40 of 57 [70.1%]), some of them with a ketogenic diet (3 of 57 [5.2%]), and some others more recently with coenzyme Q10(3 of 57 [5.2%]).

They were visited every 3 months during the first year and every 6 months thereafter. Clinical examination and general biochemical studies (hemogram, glucose, hepatic and muscle enzymes, and renal function), plasma lactate, amino acids, and urinary organic acids were performed systematically in every visit. Ophthalmo-logic and audioOphthalmo-logic evaluations and heart ultrasonography were performed every 6 months during the first year and every 6 to 12 months thereafter. Brain imaging was conducted initially and on clinical demand.

We established the following outcome categories.

Clinical Outcome Categories

1. Progression toward neurologic disease. 2. Progression toward hepatodigestive disease. 3. Progression toward myopathic disease. 4. Progression toward multisystem disease.

Biochemical Outcome Categories

1. Plasma lactate-level normalization or initially normal lactate level that remained unchanged.

2. A decrease but no normalization in plasma lactate level. 3. Persistent hyperlactacidemia.

The clinical, biochemical, enzymatic, and outcome characteris-tics of the patients are shown in Table 1.

TABLE 1. Clinical, Biochemical, Enzymatic, and Outcome Characteristics of the Patients

Initial Symptoms n Initial Lactate Level Enzymatic Defect Clinical Outcome Lactate Outcome

N M Mo S CI CI⫹CIV CIV G O Ne H M Mu NL DL HL

H⫹T⫹F 25* 3 4 18* 8* 8* 2 2 5 2 2 2 19* 7 4 14*

Neurologic 21* 3 1 3 14* 8* 3 4 2 4 7 1 13* 5 4 12*

Hypotonia 19 3 1 2 13* 8* 2 3 2 4 7 1 11* 5 4 10*

Leigh syndrome 1 1* 1* 1* 1*

Dyskinesias 1 1* 1* 1* 1*

Coma 2 1* 1* 1* 1* 1* 1* 1* 1*

Hypertonia 2 1* 1* 1* 1* 2* 2*

Seizures 1 1* 1* 1* 1*

Tremor 2 1* 1* 1* 1* 2* 1* 1*

Microcephaly 1 1* 1* 1* 1*

Hepatodigestive 15* 3 12* 7* 3 4 1 2 4 9* 3 12*

Vomiting 2 2* 1* 1* 2* 1* 1*

Liver failure 5 2 3* 3* 1 1 1 1 3* 1 4*

Hepatomegaly with high transaminases

10 2 8* 4* 2 3 1 2 8* 2 8*

Cholestasis 2 1* 1* 2* 1* 1* 2*

Hypoglycemia 9* 1 1 7* 4* 4* 1 1 2 1 5* 3 1 5*

Failure to thrive 8* 5* 3 5* 1 2 1 2 5* 1 2 5*

Cardiac 3* 1 2* 1 1* 1* 1* 1 2* 1 2*

Cardiac failure 1 1* 1* 1* 1*

Hypertrophic cardiomyopathy 2 2* 1 1* 1* 2* 2*

Hematologic 3* 1 2* 3* 1 2* 1 2*

Anemia 2 1* 1* 2* 1* 1* 1* 1*

Pancytopenia 1 1* 1* 1* 1*

Arthrogryposis 3* 2* 1 2* 1 1 2* 2* 1

Ocular (nystagmus) 2* 1* 1* 1* 1* 1* 1* 1* 1*

CPK increase 1* 1* 1* 1* 1*

Dysmorphy 1* 1* 1* 1* 1*

Initial symptoms: N indicates normal; M, mild; Mo, moderate; S, severe; G, generalized; O, others; Ne, neurological; H, hepatodigestive; M, myopathic; Mu, multisystem; NL, normalized lactate or initially normal remaining unchanged; DL, decreased lactate level; HL, persistent high lactate level; H⫹T⫹F, hypotonia, tachypnea, and failure to thrive.

Statistical Analysis

We used 2 different statistical approaches:

1. The probabilities of survival depending on the presentation symptoms, the enzymatic deficiency, and the initial lactate-level category were estimated by using the Kaplan-Meier method for univariate analysis, with the difference calculated by using a log-rank test.6_{The survival probability as a function}

of the initial total amount of lactate (mmol/L) was analyzed by using the Cox regression method.

2. We described 4 main groups of patients depending on their survival time: (1) those who died during the first 3 months of life; (2) those who died between 4 and 12 months of life; (3) those who died between 1 and 3.9 years of life; and (4) those who survived beyond 4 years of life. We compared these groups as a function of the initial lactate level and clinical presentation, enzymatic defect, and clinical and biochemical outcomes. This analysis was performed by using the␹2_test.

Finally, we attempted to relate the clinical and biochemical outcomes of these patients with the presentation symptoms initial lactate and enzymatic deficiency. Comparison between groups was performed by using the␹2_test.

We did not perform statistical analysis in function of the dif-ferent treatments because of a great heterogeneity in therapeutic trials and a short, irregular follow-up in those treated with coen-zyme Q10.

RESULTS

At the time of data collection, 33 patients had died (18 males and 15 females), 12 remained alive (6 males and 6 females), and 12 had been lost in the follow-up. Concerning the 12 patients who were lost, the mini-mum follow-up from the beginning of symptoms was 30 days (maximum: 3 years; median: 180 days). Regarding the 33 patients who died, the age of death ranged from 2 days to 3 years (median: 90 days). Most of the patients died during the first 3 months (16 of 33), 12 patients from 3 to 12 months, and 7 patients beyond 1 year of life. The 15 patients who were alive at the time of this writing had follow-up ranging from 30 days to 18 years (median: 3.2 years; mean: 5.4 years; SD: 5.4 years): 2 patients younger than 2 months, 4 patients from 1 to 3 years of age, 6 patients older than 4 years whose ages are 4, 5, and 6 years, respectively, 2 patients who are 12 years of age, and 1 who is 18 years (Fig 1).

Clinical outcome mostly tended toward a multi-system disorder (39 of 57 [68.4%]) followed by neu-rologic (9 of 57 [15.8%]), hepatodigestive (6 of 57 [10.5%]), or myopathic (3 of 57 [5.2%]) disease (see Table 2 for details).

Biochemical progression regarding blood lactate level showed persistent hyperlactacidemia in the ma-jority of patients (31 of 57 [54.4%]). Plasma lactate-level normalization and consistently normal lactate levels were found in 13 of 57 (22.8%) patients. The same number of patients had a lactate-level decrease, but it was not normalized.

Survival Studies Using the Kaplan-Meier Method We found no statistical differences in survival probability depending on the diverse clinical presen-tation forms (combined, neurologic, hepatic, failure to thrive, cardiac, or multiorganic), the different

ini-Fig 1. Distribution of patients according to the survival age. The x-axis represents the different survival age groups, and the y-axis represents the number of patients. Survival groups are defined as follows: 0, 0 to 3 months, 24 patients (41.4%); 1, 4 to 12 months, 15 patients (25.9%); 2, 1 to 3.9 years, 10 patients (17.2%); 3, beyond 4 years, 8 patients (13.8%). Patients surviving up to 3 months are the most numerous.

TABLE 2. Details of Clinical Outcome

Outcome Type Clinical Expression Patients,n/N(%) Most Prevalent Disorders

Multiorganic Neurologic 38/39 (97.4) Mental retardation, Leigh syndrome, regression,

seizures, abnormal movements

Hepatic 26/39 (66.6) Hepatomegaly with elevated transaminases, liver

failure

Failure to thrive 21/39 (53.8) Growth below⫺3 SD, anorexia

Cardiac 12/39 (30.7) Hypertrophic and dilated cardiomyopathy, heart

failure

Myopathic 6/39 (15.1) Atrophy, elevated CPK, weakness

Neurosensorial 5/39 (12.8) Nystagmus, poor vision, cataracts, pigmentary

retinitis, deafness

Renal 4/39 (10.2) Proximal tubulopathy

Neurologic Severe encephalopathy 4/9 (44.4) Severe mental retardation, spasticity, seizures

Psychomotor delay and abnormal movements

2/9 (22.2) Moderate to mild psychomotor delay and dyskinesias (dystonias, myoclonias)

Psychomotor delay 2/9 (22.2) Moderate to mild isolated psychomotor delay

Leigh syndrome 1/9 (11.1)

Hepatic Cirrhosis 3/6 (50.0)

Liver failure 2/6 (33.3)

Hepatomegaly 1/6 (16.6)

Myopathic Elevated CPK 1/2 (50.0)

tial lactate-level categories (normal, mild, moderate, or severe), or the diverse enzymatic deficiencies. We also did not find differences in survival probability depending on the blood lactate-level presentation values (significance of P ⫽ .6241) after the Cox re-gression test. However, we observed a common ten-dency in all the survival figures (Fig 2): patients who survived beyond 2.5 to 3 years were more likely to remain alive for a long period of time regardless of the initial symptoms, lactate level, or enzymatic de-ficiency.

Relation of Clinical and Biochemical Outcome With the Initial Clinical and Biochemical Characteristics Using

␹2_Test

Survival-Time Category

We found statistical significance when comparing survival-time categories with lactate-level outcome (P⬍ .0005; 2-sided) (Table 3 and Fig 3A). In fact, 21 (87.5%) of 24 patients with persistent hyperlactaci-demia died during the first 3 months of life. On the other hand, we also found a positive correlation (P⫽ .0019; 2-sided) with enzymatic deficiency: those pa-tients with CI⫹CIV deficit and those with general-ized deficiency had a clearly shorter survival time (most of them [11 of 12 and 4 of 7 patients, respec-tively] died during the first 3 months of life).

Initial Presentation

Initial Symptoms and Clinical Outcome (Comparison Between the Specific Type of Initial Symptoms and the Categories of Clinical Outcome as Described in “Patients and Methods”)

Although most of the patients developed a multi-system disorder as they evolved, we could find sta-tistical significance (P ⫽ .0017; 2-sided) because of patients later progressing toward a neurologic dis-ease: most of these patients presented initially with symptoms of nervous system involvement (7 of 9).

Initial Symptoms and Lactate Outcome

In general, a notable proportion of the patients had persistent hyperlactacidemia, but we only found clear statistical significance in those who presented with hepatic symptoms (P ⫽ .028;2-sided). These newborns always had hyperlactacidemia that was severe and persistent in the majority of the cases (80%) (Fig 3B).

Initial Lactate Level and Clinical Outcome

We found a high proportion of patients (73.6%) presenting with severe hyperlactacidemia and later developing a multiorganic disorder. This finding has a positive statistical significance (P⫽.0048; 2-sided) (Fig 3C).

Initial Lactate Level and Lactate Outcome

An important statistical significance (P ⬍ .0005; 2-sided) was found. In fact, we observed that the lower the initial lactate level, the more the chances of normalizing or decreasing it. On the contrary, severe hyperlactacidemias were more likely to remain at these very high levels.

Enzymatic Deficiency and Clinical Outcome

Patients affected with CI deficiency (82.61%) and CI⫹IV (83.33%) were inclined to develop multisys-tem involvement. These results have statistical sig-nificance (P ⫽.0006; 2-sided) (Fig 3D).

Enzymatic Deficiency and Lactate Outcome

We did not find statistical significance between these 2 variables (P⫽ .0290; 2-sided).

Description of the Patients Surviving⬎4 Years

Among patients with evolution to neurologic dis-ease (4, 6, and 18 years of age at the time of this writing), 2 patients (Table 1, patients 3 and 6) pre-sented with severe liver failure, but hepatic function

normalized over time (at 9 months and 2 years, re-spectively) (Table 4). In patient 6 the diagnosis was possible at 17 years of age after a liver biopsy despite the absence of hepatic involvement over a period of 15 years. In this case, a muscle biopsy performed at 2 years of life was normal.

Regarding patients with evolution to myopathic disease (each 12 years of age, patients 4 and 5), the presentation was severe and life threatening in pa-tient 4 (CI [14 mU/units of citrate synthase; range: 70 –250]⫹CIV [200 mU/units of citrate synthase; range: 810 –3120]); furthermore, multiorganic in-volvement was present during the first 2 years of life, but it gradually normalized, giving rise to myopathic dysfunction only at 3 years of age. At 9 years, elec-tromyography and nerve-conduction studies dis-closed signs of anterior horn involvement. This pa-tient now has mild proximal limb weakness (3 of 5) with autonomous motor function and exercise intol-erance, slightly elevated muscle enzymes (creatine phosphokinase [CPK]: 245 IU/L; lactate dehydroge-nase: 1272 U/L), and growth retardation (height and weight are at⫺2.5 SD), but brain MRI and cognitive function are normal. Patient 5 presented with feed-ing difficulties, tachypnea, and hypotonia. Marked limb hypotonia with weak deep reflexes were present from the first months of life. From the first years of life, limb weakness as well as mild bilateral ptosis were present. At 5 years, muscular power was

⫹3/5 in proximal limb muscles. Now, the boy is 12 years old, he has normal intelligence and myopathic face with persistent muscle weakness (⫹3– 4/5) but autonomous gait, Gower’s sign, and generalized slowness in physical activities. Delayed bone age and growth-hormone deficiency have also been found.

DISCUSSION

This is the first study showing data on the long-term outcome of various types of neonatal-onset mi-tochondrial cytopathies (MCs). Some reports have described the outcome of isolated cases,7–10_{but we}

have not come across studies of large series in the medical literature, nor have we found a systematic study concerning survival or potential risk factors that could modify the natural course of this type of metabolic disorder starting in infancy. By contrast, data dealing with clinical presentation have been widely reported.1–3,11–13 _{There are several factors}

that could explain this lack of information. The high early mortality and the important number of patients

that are lost in follow-up are probably the most relevant limiting conditions.

In this series we analyzed different outcome as-pects of these patients by using diverse approaches: general description, statistical methods, and detailed medical reports of the oldest surviving patients.

A global overview of this work would highlight some main points. First, found a remarkable mortal-ity rate (33 of 57 [57.8%]) that is especially high in the first 3 months of life (16 of 33 [48.4%]). This finding supports the widespread perception of such disor-ders as being fatal in the newborn.4 _{Of our 57}

pa-tients, 12 (21%) were lost in follow-up; this fact im-plies that some censored cases (incomplete observations; the final situation of the patient cannot be evaluated) will constantly appear in our survival statistical analysis. Despite this limitation, most of these patients had been followed for a relatively long period of time, and their information is useful to the results of our study. The other 21% corresponds to the surviving children. Half of them (10.5%) are now older than 4 years (mean: 9.5 years), and 3 are in their second decade of life; among them, only 2 are cog-nitively normal.

Second, we observed a clear tendency for devel-oping multiple health handicaps in a notable propor-tion of the patients (68%). Therefore, not only is survival expectance low, but the chances of good outcome are very poor.

Finally, the third general trait that is worth of consideration is the great proportion of patients with persistent high lactate level over time (31 patients with persistent unchanged hyperlactacidemia ⫹ 13 patients with a decrease but not normalization in plasma lactate level: 44 of 57 [77.1%]). This fact makes a remarkable difference compared with late-onset mitochondrial disorders, in which lactate level is lower and tends to normalize in a high proportion of patients.14,15

We analyzed these general main features by using different statistical methods to establish possible prognostic factors contributing to the natural course of these disorders.

Survival studies did not disclose significant differ-ences with any selected variable (clinical presenta-tion form, initial lactate level, or enzymatic defi-ciency). By contrast, we could demonstrate statistical significance when comparing survival-time catego-ries with different variables. This apparent contra-diction is a result of the difficulties in analyzing a

TABLE 3. Summary of Statistical Analysis (␹2_Test)

Survival Time,P

Clinical Outcome,P

Lactate Outcome,P

Initial Lactate,P

Initial Symptoms,P

Enzymatic Defect,P

Survival time .256 ⬍.0005 .059 .384* .019

Clinical outcome .256 .015 .048 .017 (NRL) .006

Lactate outcome ⬍.0005 .015 ⬍.0005 .028 (HEP) .290

Initial lactate level .059 .048 ⬍.0005 .333* .224

Initial symptoms .384 .017 (NRL) .028 (HEP) .333 .384

Enzymatic defect .019 .006 .290 .224 .384*

Comparison between categorical parameters (cross tabs,␹2_{test). All these parameters are composed of several categories and are defined}

in the text. NRL indicates neurological symptoms; HEP, hepatic symptoms

population with a great age dispersion (from 0 to 18 years at the time of data collection) and 12% of censored cases (patients who were lost in follow-up) by using survival Kaplan-Meier curves. Therefore, this test was not useful for identifying risk factors in our study; however, we could observe that all the survival curves marked a break point around 2.5 to 3 years. In general, children who survived this “criti-cal” age had more probability of surviving for a long period of time. This finding may reflect the strong sensitivity of very young children to an impaired ATP production in a period of time at which energy demands are very high.

We identified different prognostic factors relating

initial and follow-up variables. Concerning clinical presentation, we identified neurologic and hepatic symptoms as potential prognostic outcome factors. In fact, in newborns with initial neurologic signs we detected a strong risk to develop an established neu-rologic illness over time. No other presentation type could statistically predict the clinical outcome. This finding is logical if we consider that many other neonatal symptoms are more likely to regress, be-cause they represent an age-linked manifestation rather than features exclusively caused by the dis-ease. Furthermore, the higher division rate of hetero-plasmic cells in other organs can minimize the clin-ical expression as time passes.16 _{Obviously, this}

Fig 3. Relation of clinical and biochemical outcome with initial clinical and biochemical characteristics. The y-axis represents the number of patients. A, Relation between lactate outcome and survival groups (P⬍.0005);Pvalues reflect the significance between the outcome of plasma lactate level and the different survival age groups. There is statistical significance because of the persistent lactate elevation in patients who did not survive beyond 3 months of life. B, Relation between lactate outcome and initial hepatic symptoms (P⫽.028);P

cannot be the case for the nervous system. On the other hand, initial hepatic symptoms represented a risk factor of persistent severe hyperlactacidemia. Reduced activity of respiratory chain complexes has been associated with liver disease of varying sever-ity; however, neonatal presentation predominates, and one of the key features to be noted is severe lactic acidemia.17–19 _{CI, CIII, and CIV as well as}

mito-chondrial DNA-depletion syndrome are related to neonatal liver failure presenting with severe hyper-lactacidemia.17–25 _{In our series, CI deficiency and}

mitochondrial DNA depletion were the most fre-quent deficiencies.

Regarding initial lactate level, the group of chil-dren who did not survive the first 3 months of life had a clear, more persistent and severe hyperlactaci-demia with significant differences in relation to the other survival age groups. It is interesting to note that more than one third of patients who die between 4 and 12 months die despite normalizing their blood lactate levels. On the other hand, initial severe hy-perlactacidemia indicated a significantly high pro-pensity to develop a multiorganic disease. Therefore, initial lactic acidosis in neonatal MCs resulted in a bad prognosis sign, not only because of the shorter survival expectance but also because of the severity of the disease involving several organs.

If we consider the different type of enzymatic ab-normalities, a significantly higher mortality was ob-served in combined CI⫹CIV forms and generalized deficiencies in very young children (0 –3 months). The severity of generalized deficiencies is well known in different reported cases, above all if they are caused by mitochondrial DNA depletion.26–28_It

also has been reported that patients with CI defi-ciency combined with other defects tend to have a more severe illness when compared with those with isolated CI.29 _{A fatal infantile multisystemic form of}

CI deficiency has been described in different arti-cles.30,31 _{Although many cases of fatal CI were}

present in our population, this group did not show statistical differences compared with the others.

CI and CI⫹IV deficiencies in the total population represented, in our study, a negative outcome factor because they evolved toward a multisystemic dis-ease in a high proportion of patients. Nevertheless, this finding could be explained by the important prevalence of these deficiencies in our series associ-ated with the well-known general tendency to de-velop multiorganic involvement in mitochondrial disorders.

Concerning the 6 oldest surviving patients, only 2 could be considered as having “benign” or “mild” forms because of their normal intellectual level and minor motor limitations without dysfunction of other organs. Other than limitation in some physical activities, these patients can lead a normal life. They are similar in some aspects to the described cases of those with benign reversible muscle cytochrome c oxidase deficiency32–34_{but with some discrepancies.}

In our patients, different degrees of CI deficiency are associated. The great majority of benign mitochon-drial myopathies have been related to cytochrome c oxidase deficiency; however, at least 1 case of com-bined CI and CIII has been described.35_Furthermore,

in 1 of them signs of spinal anterior horn involve-ment are present. Mitochondrial myopathies mim-icking early spinal progressive muscular atrophies, although very rare, have also been reported.36,37

Sev-eral elements, however, are atypical in our case, such as the late onset and the mild nonprogressive clinical expression.

Regarding the other 4 patients, 1 has a multior-ganic disease with persistent severe lactic acidosis, despite which she has a preserved quality of life. This is exceptional, because both features (persistent se-vere hyperlactacidemia and multiorganic involve-ment) are theoretically bad prognostic factors. Two patients evolved toward a more common clinical expression in these disorders: neurologic disease manifested as slight-moderate mental retardation and different motor handicaps. Finally, our older patient has a severe encephalopathy. Some aspects of this late case are worthy of consideration. The first is the normalization of severe initial liver failure. We observed a similar phenomenon in other patients in our series. A progressive reversion of liver failure and lactic acidosis was reported previously in a child with liver and mitochondrial DNA depletion38 _and

in one with cytochrome c oxidase deficiency.39

Sec-ond, although liver function tended to normalize over time and the expression of the disease is now exclusively neurologic, the defect was only found in hepatic tissue.

CONCLUSIONS

Neonatal MCs have a high early mortality and many chances to develop complex multiorganic dis-eases or severe encephalopathies over time. Persis-tent lactic acidosis and combined enzymatic deficien-cies are markers of poor prognosis. Some initial symptoms can improve gradually and later

normal-TABLE 4. Characteristics of Patients Surviving⬎4 Years Patient No. Current Age, y Gender Enzymatic Defect Initial Symptom Initial Lactate Level, mmol/L

Clinical Outcome Current Lactate

Level, mmol/L

1 4 Female CIII⫹CIV Hypotonia 1.8 Moderate MR, hypoacusia, basal ganglia

calcifications, dystonia

Normal

2 5 Female Generalized T/H/F 15.2 Slight MR, tubulopathy, cardiomyopathy 8–10

3 6 Female CI Liver disease 9.3 Slight MR Normal

4 12 Female CI⫹CIV T/H/F 6.1 Muscle weakness, spinal atrophy-like

findings, normal IQ

Normal

5 12 Female CIV T/H/F 1.7 Muscle weakness, normal IQ Normal

6 18 Male CIV liver Liver disease 7.6 Severe encephalopathy dystonia Normal

ize, giving rise to less severe forms that, in some exceptional cases, evolve to “benign” forms. Exclu-sively myopathic outcome is the most frequent “be-nign” clinical type.

ACKNOWLEDGMENT

We thank Susana Ochoa from the Fundacio´ Sant Joan de De´u for support with the statistical analysis.

REFERENCES

1. Munnich A, Rotig A, Chretien D, et al. Clinical presentation of mito-chondrial disorders in childhood.J Inherit Metab Dis.1996;19:521–527 2. Nissenkorn A, Zeharia A, Lev D, et al. Multiple presentation of

mito-chondrial disorders.Arch Dis Child.1999;81:209 –215

3. DiMauro S, Schon EA. Mitochondrial respiratory-chain diseases.N Engl J Med.2003;348:2656 –2668

4. Sue CM, Hirano M, DiMauro S, De Vivo DC. Neonatal presentation of mitochondrial metabolic disorders.Semin Perinatol.1999;23:113–124 5. Poggi-Travert F, Martin D, Billette de Villemeur T, et al. Metabolic

intermediates in lactic acidosis: compounds, samples and interpreta-tion.J Inherit Metab Dis.1996;19:478 – 488

6. Kaplan EL, Meier P. Nonparametric estimation from incomplete obser-vations.J Am Stat Assoc.1958;53:457– 481

7. Ishikawa Y, Goto Y, Ishikawa Y, Minami R. Progression in a case of Kearns-Sayre syndrome.J Child Neurol.2000;15:750 –755

8. Finsterer J, Stollberger C, Blazek G, Spahits E. Cardiac involvement in myotonic dystrophy, Becker muscular dystrophy and mitochondrial myopathy: a five-year follow-up.Can J Cardiol.2001;17:1061–1069 9. Shoubridge EA. Cytochrome c oxidase deficiency. Am J Med Genet.

2001;106:46 –52

10. Triepels RH, Van Den Heuvel LP, Trijbels JM, Smeitink JA. Respiratory chain complex I deficiency.Am J Med Genet.2001;106:37– 45 11. DiMauro S, Andreu AL, De Vivo DC. Mitochondrial disorders.J Child

Neurol.2002;17(suppl 3):S35–S45; discussion S46 –S47

12. Smeitink JAM. Mitochondrial disorders: clinical presentation and diag-nostic dilemmas.J Inherit Metab Dis.2003;26:199 –207

13. Skladal D, Sudmeler C, Konstantopoulou V, et al.Clin Pediatr (Phila). 2003;42:703–710

14. Zeviani M, Bertagnolio B, Uziel G. Neurologic presentations of mito-chondrial diseases.J Inherit Metab Dis.1996;19:504 –520

15. Sciacco M, Prelle A, Comi G, et al. Retrospective study of a large population of patients affected with mitochondrial disorders: clinical, morphological and molecular genetic evaluation.J Neurol.2001;248: 778 –788

16. Andreu A, DiMauro S. Current classification of mitochondrial disor-ders.J Neurol.2003;250:1403–1406

17. Ronald J, Sokol RJ, Treem WR. Mitochondria and childhood liver dis-eases.J Pediatr Gastroenterol Nutr.1999;28:4 –16

18. Vilaseca MA, Briones P, Ribes A, Carreras E, Llacer A, Querol J. Fatal hepatic failure with lactic acidaemia, Fanconi syndrome and defective activity of succinate:cytochrome c reductase.J Inherit Metab Dis.1991; 14:285–288

19. Treem WR, Sokol RJ. Disorders of mitochondria.Semin Liver Dis.1998; 18:237–254

20. Bioulac-Sage P, Parrot-Roulaud F, Mazat JP, et al. Fatal neonatal liver failure and mitochondrial cytopathy (oxidative phosphorylation

deficiency): a light and electron microscopic study of the liver. Hepatol-ogy.1993;18:839 – 846

21. Cormier-Daire V, Chretien D, Rustin P, et al. Neonatal and delayed-onset liver involvement in disorders of oxidative phosphorylation.J Pe-diatr.1997;130:417– 422

22. Mazziotta MRM, Ricci E, Bertini E, et al. Fatal infantile liver failure associated with mitochondrial DNA depletion. J Pediatr.1992;121: 896 –901

23. Bakker HD, Scholte HR, Dingemans KP, Spelbrink JN, Wijburg FA, Van den Bogert C. Depletion of mitochondrial DNA in a family with fatal neonatal liver disease.J Pediatr.1996;128:683– 687

24. Maaswinkel-Mooij PD, Van den Bogert C, Scholte HR, Onkenhout W, Brederoo P, Poorthuis BJ. Depletion of mitochondrial DNA in the liver of a patient with lactic acidemia and hypoketotic hypoglycemia.J Pe-diatr.1996;128:679 – 682

25. Leshinsky-Silver E, Levine A, Nissenkorn A, et al. Neonatal liver failure and Leigh syndrome possibly due to CoQ-responsive OXPHOS defi-ciency.Mol Genet Metab.2003;79:288 –293

26. Rabinowitzs SS, Gelfond D, Chen CK, et al. Hepatocerebral mitochon-drial DNA depletion syndrome: clinical and morphologic features of a nuclear gene mutation.J Pediatr Gastroenterol Nutr.2004;38:216 –220 27. Carrozzo R, Bornstein B, Lucioli S, et al. Mutation analysis in 16 patients

with mtDNA depletion.Hum Mutat.2003;21:453– 454

28. Elpeleg O, Mandel H, Saada A. Depletion of the other genome-mitochondrial DNA depletion syndromes in humans.J Mol Med.2002; 80:389 –396

29. Korenge G, Bentlage H, Ruitenbeek W, et al. Isolated and combined deficiencies of NADH dehydrogenase (complex I) in muscle tissues of children with mitochondrial myopathies.Eur J Pediatr.1990;150:104 –108 30. Loeffen JL, Smeitink JA, Trijbels JM, et al. Isolated complex I deficiency in children: clinical, biochemical and genetic aspects.Hum Mutat.2000; 15:123–134

31. Kirby DM, Crawford M, Cleary MA, Dahl HH, Dennett X, Thorburn DR. Respiratory chain complex I deficiency: an underdiagnosed energy generation disorder.Neurology.1999;52:1255–1264

32. DiMauro S, Nicholson J, Hays A, et al. Benign infantile mitochondrial myopathy due to reversible cytochrome c oxidase deficiency.Ann Neu-rol.1983;14:226 –234

33. Zeviani M, Koga Y, Shikura K, et al. Benign reversible muscle cyto-chrome c oxidase deficiency: a second case.Neurology.1987;37:64 – 67 34. Servedei S, Bertini E, Dionisi-Vici C. Benign infantile mitochondrial

myopathy due to reversible cytochrome c oxidase deficiency: a third case.Clin Neuropathol.1988;7:209 –210

35. Castro-Gago M, Eirı´s J, Pintos E, et al. Benign congenital myopathy associated with a partial deficiency of complexes I and III of the mito-chondrial respiratory chain [in Spanish].Rev Neurol.2000;31:838 – 841 36. Pons R, Andreetta F, Wang CH, et al. Mitochondrial myopathy

simu-lating spinal muscular atrophy.Pediatr Neurol.1996;15:153–158 37. Salo MK, Rapola J, Somer H, et al. Reversible mitochondrial myopathy

with cytochrome c oxidase deficiency.Arch Dis Child.1992;67:1033–1035 38. Ducluzeau PH, Lachaux A, Bouvier R, et al. Progressive reversion of clinical and molecular phenotype in a child with liver mitochondrial DNA depletion.J Hepatol.2002;36:698 –703

DOI: 10.1542/peds.2004-2407

2005;116;1170

Pediatrics

Saudubray

A. García-Cazorla, P. De Lonlay, M.C. Nassogne, P. Rustin, G. Touati and J.M.

Patients

Long-term Follow-up of Neonatal Mitochondrial Cytopathies: A Study of 57

Services

Updated Information &

http://pediatrics.aappublications.org/content/116/5/1170 including high resolution figures, can be found at:

References

http://pediatrics.aappublications.org/content/116/5/1170#BIBL This article cites 38 articles, 4 of which you can access for free at:

Subspecialty Collections

sub

http://www.aappublications.org/cgi/collection/fetus:newborn_infant_

Fetus/Newborn Infant

following collection(s):

This article, along with others on similar topics, appears in the

Permissions & Licensing

http://www.aappublications.org/site/misc/Permissions.xhtml in its entirety can be found online at:

Information about reproducing this article in parts (figures, tables) or

Reprints

DOI: 10.1542/peds.2004-2407

2005;116;1170

Pediatrics

Saudubray

A. García-Cazorla, P. De Lonlay, M.C. Nassogne, P. Rustin, G. Touati and J.M.

Patients

Long-term Follow-up of Neonatal Mitochondrial Cytopathies: A Study of 57

http://pediatrics.aappublications.org/content/116/5/1170

located on the World Wide Web at:

The online version of this article, along with updated information and services, is

by the American Academy of Pediatrics. All rights reserved. Print ISSN: 1073-0397.