Predictors of delay in the diagnosis and clinical trial entry of amyotrophic
lateral sclerosis patients: A population-based study
Stefano Zoccolellaa, Ettore Beghib,c, Guerrino Palaganoa, Angela Fraddosioa, Vito Samarellia,
Paolo Lambertia, Vito Leporea, Luigi Serlengad, Giancarlo Logroscinoe,⁎,
for the SLAP registry
aDepartment of Neurological Sciences, University of Bari, Italy
bIstituto Ricerche Farmacologiche Mario Negri, Milano, Italy
cClinica Neurologica, Università di Milano–Bicocca, Monza, Ospedale “S. Gerardo”, Monza, Italy
dOperative Unit of Neurology, Andria (BA), Italy
eDepartment of Epidemiology HSPH 3-819 Harvard University, 677 Huntington Avenue,
Boston, MA, 02115, United States
Received 10 October 2005; received in revised form 5 June 2006; accepted 28 June 2006
Available online 22 August 2006
Background: The El Escorial and the revised Airlie House diagnostic criteria for amyotrophic lateral sclerosis (ALS) were introduced to select
patients for clinical trials. Heterogeneity of clinical presentation at onset and delay in diagnosis may decrease the likelihood for trial entry.
Objective: Identify risk factors for delay in the diagnosis and trial exclusion.
Methods: ALS incident cases were identified with El Escorial (EEC) and Airlie House criteria (AHC) through a population-based registry
established in Puglia, Southern Italy, in the years 1998–99.
Results: 130 ALS incident cases were diagnosed with a median interval between onset of symptoms and diagnosis of 9.3 months and not
different across both EEC and AHC categories. Twenty percent of cases were not eligible for clinical trials according to the AHC. About 5%
of subjects in this series died with only lower motor neuron signs. Predictors for delay in the diagnosis were age between 65 and 75 years and
spinal onset while fasciculations and cramps as first symptoms were predictors of exclusion from trials.
Conclusions: In this population-based series, diagnostic delay was longer in subjects with spinal onset and age between 65 and 75 and
fasciculation as first symptoms. About 80% of incident cases were trial eligible with AHC criteria. However, a significant number of subjects
with ALS, characterized by a limited spread of signs, were not trial eligible while alive.
© 2006 Elsevier B.V. All rights reserved.
Keywords: Amyotrophic lateral sclerosis; Incidence; El Escorial criteria; Airlie House criteria; Diagnostic delay; Clinical trials
The diagnosis of ALS is made on clinical grounds, because
of the absence of a biological marker and of specific
neuroradiological or neurophysiological features . The
clinical presentation of ALS is extremely variable and the
can delay the time to diagnosis  and decrease the likelihood
Patients recruited in population-based studies differ from
patients recruited in ALS tertiary centers, because of several
clinical characteristics including age at onset, frequency of
bulbar ALS, and sex distribution [4–12] and consequently
probably have a different likelihood to be included in clinical
trials. However, only a few reports from population-based
studies on trial eligibility according to EEC and AHC have
been published [9–11]. Their results indicate that a high
Journal of the Neurological Sciences 250 (2006) 45–49
⁎Corresponding author. Tel.: +1 617 432 2652; fax: +1 617 566 7805.
E-mail address: email@example.com (G. Logroscino).
0022-510X/$ - see front matter © 2006 Elsevier B.V. All rights reserved.
number of patients would not be suitable for trial entry at the
the disease as well (10–20%) [8,9,13].
The aims of the study are to identify any factors that
determine the delay in the diagnosis and the exclusion from
2. Materials and methods
The source of data for this study is a population-based
registry of newly diagnosed ALS cases that was established
in Puglia, Southern Italy, in 1997 . The surveillance
began on January 1, 1998. We used five different sources for
1) The SLAP registry (Sclerosi Laterale Amiotrofica-
Puglia), a network of 23 neurological and neurophysio-
logical facilities in the region .
2) The Hospital Discharge Diagnosis Data Bank, a regional
computerized system including the diagnosis of each
hospitalized patient in Puglia, the area of the registry. This
system is inclusive of everyALSinpatient,bothfrom public
and private clinics (including nursing homes).
3) Although the diagnosis of ALS without hospitalization is
very unlikely in Italy, we checked all facilities where EMG
is performed in the region to also intercept possible ALS
outpatients that did not go through hospitalization.
has a roster of ALS patients in the region, as well as of the
personnel involved in the diagnosis and care.
5) The regional registry for riluzole prescription.
One neurologist is in charge of recruitment of all newly
diagnosed ALS patients for the registry in each Neurological
based on the El Escorial criteria (EEC)  and their revised
version, the Airlie House criteria (AHC) , which were used
to reclassify possible ALS cases based on EMG findings as
Clinically possible ALS and Laboratory-supported probable
ALS (PrLS-ALS). Cases were classified as PrLS-ALS when
clinical signs of UMN and LMN dysfunction were present in
region, and LMN signs defined by EMG criteria were present
in at least two limbs .
Patients with all clinical forms of ALS [including
progressive bulbar palsy (PBP), progressive muscular atrophy
(PMA) and primary lateral sclerosis (PLS)] were therefore
enrolled in the registry. Individuals under the age of 18 were
excluded to avoid misclassification of other motor neuron
diseases of genetic origin mimicking ALS. Date of onset was
based on information given by the patient while date of
diagnosis was based on review of medical charts.
All the forms were reviewed by one of the authors (L.S.)
who confirmed the diagnosis and decided the patient's
inclusion in the registry, as well as the classification based on
EEC and AHC. The forms were reviewed by the clinical
committee only if there was a conflicting diagnosis between
the site investigator and the clinical principal investigator.
Using this multisource registry, we identified all residents
18 years of age or older in Puglia diagnosed with ALS in the
2-year period from January 1, 1998 to December 31, 1999.
All patients were asked to give informed consent to
participation in the study; data were stored in a centralized
database and separate anonymous files were kept for each
Differences were tested with t-test for continuous variables
and chi-square for categorical data. Analysis of variance was
used to determine the effect of clinical variables on interval
onset-diagnosis while trial eligibility was analyzed as a
dichotomous outcome with multivariate logistic regression.
During the two-year surveillance period we identified
130 ALS patients, with a male/female ratio of 1.6. Median
age at diagnosis was 64.6 years overall, 65.4 for males (range
patients were under 45 years (7.7%). No cases were found
among subjects older than 80years. Thelimbpresentationwas
the most common: 35% of thecasespresented withupperlimb
onset, 34% with lower limb onset and 5% both in upper and
19% of all subjects (24% of women and 15% of men); and 7%
had a generalized onset (i.e. combined bulbar and limb onset).
3.1. Delay in the diagnosis
to diagnosis interval, ODI) was 9.3 months. This interval was
shorter (7 months; 3.8–49.2) for subjects with bulbar onset
compared to subjects with limb onset (10 months; 1–70.7).
The ODI was longer in males (11.2 months) than in females
(8.5 months), though the difference was not significant.
(b45 years, 7 months; 55–64 years, 8 months, ≥75 years,
7 months; p=0.006).
The ODI were not significantly different across all EEC
diagnostic groups. Patients with an ODIb6 months (n=28)
were more likely diagnosed as definite or probable ALS
(60%), whereas patients with a delayed diagnosis
(ODIN12months;n=49) morefrequentlypresented a clinical
feature of possible or suspected ALS (54%). Similar findings
were found using the cut-off value of 18 months.
In addition, the new category introduced by the AHC, the
PrLS-ALS, presented a median ODI (11.3 months) that was
not significantly different from the clinically probable
(9.1 months) and the clinically possible ALS (9.5 months).
(more than12months) was lowerin subjectswith bulbar onset
compared to other types of presentation (RR: 0.27; CI: 0.07–
0.94), and was higher in the age group between 65 and 75,
46S. Zoccolella et al. / Journal of the Neurological Sciences 250 (2006) 45–49
compared to subjects aged less than 45 years (RR: 6.5; CI:
1.18–35.74),while gender, EEC andAHC categorieswerenot
predictors of time to diagnosis.
3.2. Classification of cases and entry in clinical trials
Most clinical trials require EEC diagnosis of definite or
probable ALS for the inclusion into clinical trials; using the
EEC 57% (n=73) of our cohort would have been considered
trial eligible at the time of first diagnosis. No case of definite
ALS was diagnosed before the age of 45 years and the
percentage of patients with definite ALS progressively
increased with age.
According to the revised AHC, subjects who are
classified as PrLS-ALS meet the criteria for trial eligibility.
Sixty-five percent of possible ALS cases based on the EEC,
and 22% of the entire cohort (18 males and 11 females), were
reclassified as PrLS-ALS, based on AHC, while 16 cases
(12%), 8 males and 8 females, remained clinically possible
ALS. The PrLS-ALS presented with clinical and EMG LMN
signs in two regions, whereas UMN signs were evident in
only one region.
Thus, when the AHC were applied, 79% (n=102) of our
patients became eligible for clinical trials; this percentage
was similar for both bulbar and spinal ALS.
Cases excluded from clinical trials (ExC) in our series
(n=28; 21%) presented a slightly older median age at
diagnosis and similar sex distribution to cases eligible for
clinical trials (InC), while the median ODI was similar in the
two groups (Table 1). Muscle cramps (chi square: 4.1;
p=0.02), as well as fasciculations (chi square: 6.4;
p=0.007), were more frequently referred as the presenting
symptom in the group of ExC. Among ExC, none presented
with a generalized type of onset or with signs both in upper
and lower limbs (versus 7 and 9% of InC). Among cases
included in clinical trials dysarthria was present in one out of
three patients (Table 1), while dysphagia did not differentiate
included from excluded.
Twelve of the 28 ExC, approximately 10% of the entire
cohort, were PMA (9 males and 3 females), with a median
age of 66 years and an ODI of 9.2 months. The onset of
symptoms was in upper limbs in 60% of the cases and at the
time of diagnosis 11 of 12 PMA cases presented LMN signs
in both bulbar and spinal regions. During follow-up six of
12 PMA cases became probable or definite ALS within one
year, while the remaining patients (5% of total) died with
only LMN signs within 3 years from the diagnosis.
In multivariate modelling fasciculation as symptom at
onset was associated with a higher risk for exclusion from
clinical trials (OR: 3.9; 95% CI: 1.3–12.0; p=0.02), whereas
sex, age, other symptoms at onset and site of onset did not
significantly influenced trial inclusion in our cohort of
In this population-based study of incident ALS cases,
delay in the diagnosis was longer for subjects in the 65–
74 year old age group and with spinal onset of symptoms,
and was not different across AHC and EEC categories. If the
EEC category of definite and probable ALS were used as
inclusion criteria in clinical trials about 40% of the cases in
this series would have been excluded from the entry, while
the excluded would have been about 20% when applying the
revised criteria of AHC. The presence of fasciculations as
symptoms onset was associated with a higher risk of
exclusion from clinical trials.
After the studies in Ireland and Northern Italy [9,10] this
is the third population-based study looking at delay in the
diagnosis and trial eligibility using the AHC and EEC for
ascertainment of incident cases in a well-defined geographic
area. To our knowledge this is the first to use multivariate
analytic approach to take into account the effect in the
analysis of possible confounding factors.
4.1. Delay in the diagnosis
The median ODI observed in our study area (9.3 months)
was shorter than that observed in Northern Italy
(11.1 months) , but slightly longer when compared to
the Irish study (8 months) ; this difference could be
explained by the lower number of patients with bulbar ALS
observed in our series (24.5% of entire cohort) compared to
the Irish (approximately 40%). Bulbar onset ALS, in fact,
can be first seen at an earlier stage in the disease process,
before the spread to other regions. Access to medical care
may also determine, at least in part, this difference. However,
Trial eligibility for incident cases in a population-based registry: clinical
features at onset and diagnostic category at entry by Airlie House criteria
Eligible (n=102)Ineligible (n=28)
Median age (range)
Symptoms at onset
Muscle cramps (°)
Median ODI months (range)
Site of onset
AHC categories29 definite (28.5%)
44 probable (43%)
29 PrLS (28.5%)
16 possible (57%)
12 suspected (43%)
47S. Zoccolella et al. / Journal of the Neurological Sciences 250 (2006) 45–49
the presence of a National Health system with universal
coverage in both Ireland and Italy should minimize the role
of social factors in determining the quality and the
promptness of medical care.
The onset of symptoms in spinal regions (cervical, thoracic
and lumbo-sacral) is the other main factor to determine the
delay in the ALS diagnosis. This may be the consequence of a
more difficult differential diagnosis with an ALS mimic
No significant difference was found in ODI across the
EEC and the AHC categories; this finding confirms the
results of previous studies [9,10], indicating that the current
ALS diagnostic criteria do not distinguish subsequent stages
of the disease process, but rather reflect the heterogeneous
clinical presentation of the disease. However, subjects with a
particularly long ODI had a higher likelihood to be in the
possible and suspected categories compared to subjects with
a shorter ODI (less than 6 months). Subjects of this cohort
with a late diagnosis were therefore less likely to enter in
clinical trials. Short ODI has been indeed shown to be a poor
prognostic factor [9,10].
4.2. Classification of cases and entry in clinical trials
Using the probable and definite EEC as eligibility criteria
for clinical trials, 57% of our patients would have met the
criteria for trial eligibility, whereas almost 80% of the patients
would have been considered trial eligible at the time of
diagnosis, using the AHC. Thus, the percentage eligible for
clinical trials from this population-based series is comparable
series used for the validation of the EEC .
The PrLS-ALS represented approximately 20% of our
cases. Clinical features of these patients are characterized by
a prominent LMN involvement, documented both by clinical
and neurophysiological investigations. In a neuropatholog-
ical study  of 102 ALS cases from an autopsy series a
predominantly LMN involvement was found in 23% of the
cases. Thus, this new category may indeed correspond to the
clinical presentation of a neuropathological form with
predominant LMN involvement. PrLS-ALS cases in our
cohort of ALS patients have a clinical course and a final
outcome similar to probable ALS (log-rank: 1.5; p=0.2).
Subjects who complained of fasciculations as symptom at
onset had a much higher likelihood to be excluded from
clinicaltrials. One possibleexplanationfor these findingsmay
arise from the observation that cramps and fasciculations are
considered the result of LMN hyperactivity preceding
neuronal death . Even if cramps and fasciculations are a
frequent onset symptom, ALS patients generally do not report
these symptoms at diagnosis, or do so only retrospectively. A
phase characterized by the presence of only cramps and
fasciculations is generally very short because more disturbing
symptoms generally supervene. The subjects who, on the
contrary, are able to describe a phase characterized by
fasciculations and cramps without weakness and atrophy are
more likely to have a slower course, without upper motor
neuron involvement , and with limited spread of signs
across different body regions. It is more difficult to reach the
level of diagnostic certainty that is required to be included in
clinical trials in the presence of such clinical features. The
presence of fasciculations, particularly if widespread without
fibrillation potentials may further delay the diagnosis of ALS
at onset  and therefore the likelihood of trial entry. This
interpretation of our findings needs replication in further
The prevalence of PMA observed in our study was similar
to previous referral series from tertiary centers  and
population-based studies [8,9]. LMN clinical features with
to EEC as suspected ALS, but are not included in the revised
categories of AHC.
A prospective study on the natural history of ALS showed
that the majority of patients with LMN syndromes at an early
stage progress to other categories or have a clinical course
similar to the other ALS diagnostic categories . Neuro-
physiological and autopsy studies demonstrated that the
pyramidal tracts are often affected in patients with LMN
syndromes [18,19] and ubiquinated inclusions typical of ALS
have been found in most PMA cases in a recent report .
Familial cases carrying superoxide dismutase-1 mutations
frequently present LMN phenotype at onset [20,21]. These
cases are included in the AHC as having clinically definite
familial, laboratory-supported ALS. Therefore, at this time, it
of motoneuron diseases.
Half of the 12 PMA cases evolved into definite or probable
ALS. These findings are consistent with the hypothesis of a
clinico-pathological continuum ranging from slowly progres-
sive PMA, PMA with rapid progression to ALS to “typical”
ALS, sharing a common molecular pathology and in which the
clinical features of the single patient depend on the individual
Therefore, the follow-up does not necessarily solve the
issue of the limited spread of signs: in a report from the
population-based registry in Ireland, 10% of the patients died
without reaching trial eligibility .
This population-based study demonstrates that the use of
the AHC as eligibility criteria for clinical trials increases
enrollment to 80% compared to 57% with use of EEC.
However, a considerable number of patients (20%) do not
reach trial eligibility. The issue of limited access to clinical
trials of subjects with limited spread of signs should be
evaluated in future studies.
We thank Mr. Camillo Colapinto who was the first to
launch the idea of a regional registry for ALS in Puglia and
who always gave us great support with enthusiasm and new
ideas both in person and through the AILSA.
48S. Zoccolella et al. / Journal of the Neurological Sciences 250 (2006) 45–49
Appendix A. Sclerosi Laterale Amiotrofica-Puglia
Principal investigators: Giancarlo Logroscino (Boston);
Luigi Selenga (Andria).
Scientific committee: Ettore Beghi, Vito Lepore, Paolo
Livrea, Giancarlo Logroscino, Isabella L. Simone, Luigi
Serlenga. Clinical committee: Paolo Lamberti, Bruno
Maggio, Bruno Passarella, Vito Santamato, Luigi Serlenga,
Isabella Simone, Pasquale Simone, Franco Valluzzi. Epide-
miologic and data management unit: Vito Lepore, Saverio
Staffieri. Study monitors: Angela Fraddosio, Rino Pala-
gano, Stefano Zoccolella.
SLAP neurologists: GiuseppeBelfiore(Lecce),Giuseppe
Benedetto (Noci), Nicola Cacudi (S. Paolo, Bari), Antonio
Cazzato (Lecce), Pasquale Colamartino (Bisceglie), Pietro Di
Viesti (S. Giovanni Rotondo), Silvana Epifani (Galatina),
Francesco Lincesso (Taranto), Bruno Maggio (Conversano),
Vincenzo Monitillo (Cassano Murge), Angelo Moramarco
(Altamura), Antonello Nicolaci (Scorrano), Cecilia Nozzoli
(Brindisi), Sergio Pasca (Casarano), Rosaria Pulimeno
(Gallipoli), Giuseppe Russo (Grottaglie), Vito Santamato
(DiVenere, Bari), Isabella Laura Simone (Policlinico, Bari),
GiovanniStrafella (Andria), Maria Terraciano (Foggia), Paolo
Tota (Barletta), Francesco Valluzzi (Putignano).
 Brooks BR. El Escorial World Federation of Neurology criteria for the
diagnosis of amyotrophic lateral sclerosis. Subcommittee on Motor
Neuron Diseases/Amyotrophic Lateral Sclerosis of the World
Federation of Neurology Research Group on Neuromuscular Diseases
and the El Escorial “Clinical limits of amyotrophic lateral sclerosis”
workshop contributors. J Neurol Sci 1994;124:96–107 (Suppl).
 Iwasaki Y, Ikeda K, Ichikawa Y, Igarashi O, Kinoshita M. The
diagnostic interval in amyotrophic lateral sclerosis. Clin Neurol
 Brooks BR, Miller RG, Swash M, Munsat TL. El Escorial revisited:
revised criteria for the diagnosis of amyotrophic lateral sclerosis.
Amyotroph Lateral Scler Other Motor Neuron Disord 2000;1(5):293–9.
 RingelSP, MurphyJR, Alderson MK,BryanW,EnglandJD,MillerRG,
et al. The natural history of amyotrophic lateral sclerosis. Neurology
 Miller RG, Anderson Jr FA, Bradley WG, Brooks BR, Mitsumoto H,
Munsat TL, et al, ALS C.A.R.E. Study Group. The ALS patient care
database: goals, design, and early results. Neurology 2000;54(1):53–7.
 Norris F, Shepherd R, Denys EUK, Mukai E, Elias L, et al. Onset,
natural history and outcome in idiopathic adult motor neuron disease. J
Neurol Sci 1993;118(1):48–55.
 Haverkamp LJ, Appel V, Appel SH. Natural history of amyotrophic
lateral sclerosis in a database population. Validation of a scoring
system and a model for survival prediction. Brain 1995;118:707–19.
 Chancellor AM, Slattery JM, Fraser H, Swingler RJ, Holloway SM,
Warlow CP. The prognosis of adult-onset motor neuron disease: a
prospective study based on the Scottish Motor Neuron Disease
Register. J Neurol 1993;240(6):339–46.
 Traynor BJ, Codd MB, Corr B, Forde C, Frost E, Hardiman OM.
Clinical features of amyotrophic lateral sclerosis according to the El
Escorial and Airlie House diagnostic criteria: a population-based study.
Arch Neurol 2000;57(8):1171–6.
 Chio A, Mora G, Leone M, Mazzini L, Cocito D, Giordana MT, et al.
Early symptom progression rate is related to ALS outcome: a
prospective population-based study. Neurology 2002;59(1):99–103.
 Forbes RB, Colville S, Swingler RJ. The epidemiology of amyotrophic
lateral sclerosis (ALS/MND) in people aged 80 or over. Age Ageing
 del Aguila MA, Longstreth Jr WT, McGuire V, Koepsell TD, van
Belle G. Prognosis in amyotrophic lateral sclerosis: a population-
based study. Neurology 2003;60(5):813–9.
 Chaudhuri KR, Crump S, al-Sarraj S, Anderson V, Cavanagh J,
Leigh PN. The validation of El Escorial criteria for the diagnosis of
amyotrophic lateral sclerosis: a clinicopathological study. J Neurol
Sci 1995;129:11–2 (Suppl).
 Logroscino G, Beghi E, Zoccolella S, Palagano R, Fraddosio A,
Simone IL, et al. Incidence of amyotrophic lateral sclerosis in Southern
Italy: a population-based study. J Neurol Neurosurg Psychiatry
 Piao YS, Wakabayashi K, Kakita A, Yamada M, Hayashi S, Morita T,
Ikuta F, et al. Neuropathology with clinical correlations of sporadic
amyotrophic lateral sclerosis: 102 autopsy cases examined between
1962 and 2000. Brain Pathol 2003;13(1):10–22.
 de Carvalho M, Swash M. Cramps, muscle pain, and fasciculations:
not always benign? Neurology 2004;63(4):721–3.
 Rosenfeld J. Fasciculations without fibrillations: the dilemma of early
diagnosis. Amyotroph Lateral Scler Other Motor Neuron Disord
 Verma A, Bradley WG. Atypical motor neuron disease and related
motor syndromes. Semin Neurol 2001;21(2):177–87.
 Ince PG, Evans J, Knopp M, Forster G, Hamdalla HH, Wharton SB, et
al. Corticospinal tract degeneration in the progressive muscular
atrophy variant of ALS. Neurology 2003;60(8):1252–8.
 Cudkowicz ME, McKenna-Yasek D, Chen C, Hedley-Whyte ET,
Brown Jr RH. Limited corticospinal tract involvement in amyotrophic
lateral sclerosis subjects with the A4V mutation in the copper/zinc
superoxide dismutase gene. Ann Neurol 1998;43(6):703–10.
 Andersen PM, Nilsson P, Keranen ML, Forsgren L, Hagglund J,
Karlsborg M, et al. Phenotypic heterogeneity in motor neuron disease
patients with CuZn-superoxide dismutase mutations in Scandinavia.
49S. Zoccolella et al. / Journal of the Neurological Sciences 250 (2006) 45–49