myasthenia gravis - cross-cultural comparison - translations - validation studies
miastenia gravis - comparação transcultural - traduções - estudos de validação
Myasthenia gravis (MG) is a chronic autoimmune neuromuscular disorder characterized
by weakness of the skeletal muscles[1]. In approximately 85% to 90% of patients with MG, antibodies against nicotinic acetylcholine
receptors are identified in the neuromuscular junctions, with a smaller group of patients
having autoantibodies against muscle-specific kinase[2], low-density lipoprotein-related protein 4, and even against agrin[3],[4].
The incidence and prevalence rates of MG vary greatly from country to country, making
it difficult to generalize the data. Nonetheless, crude estimates are possible using
the combination of pooled estimates and the range of observed frequencies. The incidence
of all MG is: 5.3 per million person-years (C.I.: 4.41, 6.12), range: 1.7 - 21.3,
and prevalence: 77.67 cases per million (C.I.: 63.98, 94.30), range: 15 - 179[5].
The Myasthenia Gravis Foundation of America recommends that the Quantitative Myasthenia
Gravis Score (QMGS) be used in all prospective clinical trials on MG[6],[7]. The QMGS was developed by Besinger et al. in 1983 as a clinical scoring system
to assess clinical outcomes in patients with MG[8]. The original scale comprised eight items, graded from 0 to 3, where 3 is indicative
of a higher severity of impairment. In 1987, Tindall et al. modified the scale, increasing
the number of items from eight to 13 to improve the responsiveness of the scale in
evaluating outcomes of treatment interventions for MG[9],[10].
In 1992, Tindall et al.[10] used the modified QMGS to evaluate the effectiveness of a six-month treatment with
cyclosporine (5 mg/kg per body weight) compared to a placebo group. Thirty-nine patients
were randomized to the cyclosporine or the placebo group, with outcome measures evaluated
monthly. Outcome measures included: the QMGS, anti-human acetylcholine receptor antibody
titer, and dose of corticosteroid medication. At the six-month end-point, patients
in the cyclosporine group exhibited significant increases in strength (p = 0.004)
and reduction in antireceptor antibody titer (p = 0.01), compared to those in the
placebo group. Subsequently, the QMGS scale underwent further modification by Barohn
et al.[11]
In its current format, the QMGS is a 13-item measurement tool used to quantify disease
severity in patients with MG, including ocular, bulbar, respiratory, and limb function,
with a total score range of 0 to 39, where a higher score is indicative of greater
disease severity. The reliability and longitudinal validity of the QMGS have been
demonstrated in several studies[12],[13].
From a clinical perspective, the QMGS provides an easy-to-use outcome measure that
requires minimal equipment. Moreover, a change of 3.5 points in the total score is
considered a clinically meaningful improvement for patients with MG. Therefore, the
aim of this study was to perform the translation, cultural adaptation, and validation
of the QMGS to extend its application in clinical practice and research with patients
with MG in the Portuguese community of Brazil. According to our knowledge, this is
the first translation and cultural adaptation of QMGS for another language.
METHODS
A search of the literature was performed to identify previous research on the adaptation
and use of the QMGS for the Brazilian Portuguese population. The following databases
were searched using the combination keywords “Quantitative Myasthenia Gravis Score
and Portuguese validation”: Medline, PubMed, Scientific Electronic Library Online,
and Scientific and Technical Literature of Latin America and the Caribbean. No published
information was identified.
Four neuromuscular disease research centers collaborated in our study. All methods
and procedures were performed in accordance with the ethical standards established
in the 1964 Declaration of Helsinki (as revised in Hong Kong in 1989 and Edinburgh,
Scotland in 2000). Our study complied with the Regulatory Guidelines and Norms for
Research Involving Human Subjects of the National Health Board of the Brazilian Health
Ministry, issued in December 2012. Our study is part of a previously-published research
protocol[14], approved by the Human Research Ethics Committee of Nove de Julho University (Brazil)
under process no. 360.488 and registered with the World Health Organization under
Universal Trial Number (UTN) U1111-1147-7853 and the Brazilian Registry of Clinical
Trials (REBEC) RBR -7ckpdd.
Written informed consent was obtained prior to enrollment and participants were allowed
to withdraw from the study, at any time, without consequence. Our eligibility criteria
were as follows: a clinical diagnosis of MG according to the Myasthenia Gravis Foundation
of America (MGFA) criteria; 18 to 75 years of age; both sexes; clinically stable patients;
and capable of and willing to provide informed consent. Excluded from our study were
patients with other neuromuscular, cardiovascular, and respiratory diseases, episodes
of clinical decompensation of their disease in the two months prior to the study,
a prior stroke, history of mental disease, or abuse of drugs and/or alcohol. We undertook
the process of translation, cultural adaptation, and validation of the QMGS based
on previously-published standardized rules[15],[16]. A flowchart of the study is shown in [Figure 1].
Figure 1 Flowchart of the study.
First stage
The original version of the QMGS was translated from English to Brazilian Portuguese
by three bilingual translators, a physical therapist; a neurologist, and a certified
professional translator, all of whom were blinded to the purpose of our study. The
three translated versions were evaluated, compared, and merged into an initial Brazilian
Portuguese draft version.
Second stage
Two native English-speaking translators, a certified professional translator, and
a neurologist with no knowledge of the QMGS or of our research aim independently performed
a back-translation of the final draft version from Brazilian-Portuguese to English.
The back-translated English version was compared to the original English version with
adjustments made to the Brazilian-Portuguese version to correct identified discrepancies.
Third stage
A cultural adaptation of the Brazilian-Portuguese version of the QMGS, updated after
the back-translation process, was subsequently undertaken, enrolling 10 patients with
MG from the four neuromuscular disease research centers collaborating in our study.
We added the phrase “difficult to understand” as a response option in the Brazilian-Portuguese
version. Initially, we established an understanding of each item by 90% of healthcare
professionals for the Brazilian-Portuguese translation to be acceptable, with the
translation process to be repeated if this threshold was not achieved.
Five neurologists and five physiotherapists were recruited to verify the clarity of
clinical terminology for the translated QMGS version. Again, a threshold of ≥90% was
used for the translation to be deemed acceptable for clinical practice. In completing
their review, the healthcare professionals also commented on the overall clarity of
the QMGS. Ambiguous terms were discussed and replaced by alternative terms having
semantic equivalence and appropriate to the Brazilian culture. A final version of
the Brazilian Portuguese QMGS was consolidated without compromising the intended meaning
of the original version ([Figure 2]).
Figure 2 Quantitative tests for myasthenia gravis in Brazilian Portuguese.
Reliability and validity of the QMGS Brazilian Portuguese version
A sample of convenience of 30 patients, with a confirmed clinical diagnosis of MG
based on the criteria of the MGFA and confirmed by the neurologists, was recruited
to evaluate the reliability and validity of the Brazilian-Portuguese version of the
QMGS. The 10 patients who participated in Phase III, cultural adaptation, were included in the validation component of our study.
Two trained neurologists (observer A and B) evaluated the QMGS at three time points.
To assess inter-observer reliability, neurologists completed the QMGS, independently,
within a two-hour interval. To assess intra-observer reliability, observer A completed
the QMGS evaluation at one-week intervals, with interviews conducted at the same time
of day.
The concurrent validity of the Brazilian Portuguese version of the QMGS was evaluated
against the Portuguese version of the Myasthenia Gravis Composite Scale (MGC)[17]and the Portuguese version of the 15-item Myasthenia-specific Quality of Life Questionnaire
(MG-QOL 15)[18]in a group of 30 patients with MG. The same evaluator applied the MGC and QMGS. The
MGC is a 10-item scale designed to evaluate the function of patients with MG, with
the response scale for each MGC items weighted in terms of the impact of the item
on overall function. For example, ptosis scores 3 points, hip flexion weakness scores
5 points, and severity of respiratory symptoms for ventilator-dependent patients scores
9 points. The maximum possible score on the MGC is 50 points, with a higher score
reflecting a more severe disease state. The validity and reliability of the MGC has
been established, both in the United States and Europe, with a 3-point reduction in
total score being indicative of a clinically meaningful improvement in the function
of a patient with MG[19],[20].
The 15-item MG-QOL is an easy-to-use instrument that evaluates three dimensions of
health-related quality of life. The 15 items are scored on a scale of 0 to 4, with
the score on each item summed to provide the total score. A clinically meaningful
cutoff is not available for the MG-QOL. Therefore, the total score is interpreted
along a continuum, with a higher score being indicative of a perception of an increasingly
poorer quality of life. Both the MGC and MG-QOL are widely used to evaluate the physical,
social, and psychological aspects of life of patients with MG and are used to derive
an overall evaluation of the health-related quality of life in this clinical population[18],[21].
Lung function tests were performed on the day following the clinical evaluation, using
the KoKo® Sx 1000 system (nSpire Health Inc., Longmont, CO, USA), based on the guidelines of
the Brazilian Society of Pneumology[22]. Peak forced expiratory volume (measured over a 1-s expiration) and forced vital
capacity were used for analysis, regardless of the form of the curve (i.e., peak values
were not necessarily obtained from the best expiratory curves).
Handgrip strength was assessed using the JAMAR hydraulic hand dynamometer device (Lafayette
Instrument Company, Inc., Lafayette, IN, USA), which is widely used to provide a simple
measure of hand strength in clinical populations with neuromuscular diseases. The
JAMAR hand dynamometer is adjustable to the size of a patient’s hand[23]. Handgrip strength was evaluated with patients in a sitting position, with the shoulder
in 180º of forward flexion and the elbow in full extension.
Statistical analysis
Prior to data analysis, normality of data distribution was evaluated using the Shapiro-Wilk
test. For parametric data, including anthropometric and demographic measures, the
mean and standard deviation values were calculated.
Intra- and inter-observer reliability was assessed using Pearson’s correlation coefficient
and the intra-class correlation coefficient with two-way mixed effects model where
people effects are random and measures effects are fixed. The intra- and inter-observer
reliability was also evaluated for each item of the QMGS using Cohen’s kappa test,
with the following interpretation of the kappa index: < 0, no agreement; 0–0.20, slight
agreement; 0.21–0.40, fair agreement; 0.41–0.60, moderate agreement; 0.61–0.80, substantial
agreement; and 0.81–1, excellent agreement.
Spearman’s rank correlation coefficient was used to evaluate agreement between the
score on the items of the QMGS and the MGC and MG-QOL. For all analyses, a p value
< 0.01 was considered significant. Statistical analysis was performed with StatView
5.0 (SAS Institute, Cary, N.C., USA), and SPSS software (version 23.0, SPSS Inc. Chicago,
IL, USA).
RESULTS
Demographic data
Thirty-six patients (25 women), with a confirmed diagnosis of MG using the criteria
of the MGFA, were identified from the four neuromuscular clinics involved in our study:
the Neuromuscular Diseases Research department of the Universidade Federal de São
Paulo; the Department of Neurology of the Faculty of Sciences, physicians from Santa
Casa de São Paulo; the Department of Neurology of the Hospital do Servidor Público
Estadual; and the Child Neurology Clinic of the Hospital das Clínicas, Faculdade de
Medicina, Universidade de São Paulo. Six patients did not meet our eligibility criteria
and, therefore, were not enrolled in our study.
The demographic, clinical data, and classification of disease severity for our final
study groups are shown in [Table 1]. Of the 30 patients forming our study group, 24 were women, with a mean age of 47.6
± 11.4 years (range, 32–74 years), and a mean duration of illness of 11.33 ± 8.49
years. The distribution of MGFA classification was as follows: four patients, Class
I; 17 patients, Class II; eight patients, Class III; no patients in Class IV or V.
Among the patients in our study group, 96% were using cholinesterase inhibitors and
36% were using immunosuppressant drugs.
Table 1
Baseline clinical and demographic variables.
|
Variables
|
Patients (n = 30)
|
(%)
|
|
Female/Male
|
24/6
|
-
|
|
Age (years)
|
47.6 ± 11.4
|
-
|
|
Body Mass Index
|
27.5 ± 5.4
|
-
|
|
MGFA
|
|
I
|
2
|
6.6
|
|
IIA
|
17
|
56.6
|
|
IIB
|
2
|
6.6
|
|
IIIA
|
5
|
16.6
|
|
IIIB
|
4
|
13.3
|
|
QMGS
|
11.4 ± 5.7
|
-
|
|
MGC
|
12.93 ± 7
|
-
|
|
MG QOL – 15 score
|
21.6 ± 12
|
-
|
Note: Values shown are number (%), or mean ± SD. MGFA: Myasthenia Gravis Foundation
of America Clinical classification. QMGS: Quantitative Myasthenia Gravis Score; MGC:
Myasthenia Gravis Composite Scale; MG-QOL: Questionnaire of Life Quality Specific
for Myasthenia Gravis – 15 items.
On average, the QMGS was completed in 32 minutes, with an additional 18 minutes required
for assessment of forced vital capacity by spirometry and eight minutes for hand grip
strength. The kappa index of inter- and intra-observer reliability for each item of
the QMGS is reported in [Table 2], with all kappa values ≥0.81, indicative of an excellent reliability for all items.
Overall inter- and intra-observer reliability was confirmed by the intra-class correlation
coefficient and Pearson’s correlation coefficient reported in [Table 3]. Correlation between the QMGS and MGC was very strong (R = 0.928; p < 0.001) and
substantial between the QMGS and MG-QOL15 (R = 0.737; p < 0.001) as shown in [Figure 3].
Table 2
Assessment of intra- and inter-interviewer reproducibility of Quantitative Myasthenia
Gravis score (QMGS) with Cohen’s kappa.
|
QMGS items
|
Intra-interviewer
|
Inter-interviewer
|
|
Double vision
|
0.896
|
0.897
|
|
Ptosis
|
0.890
|
0.721
|
|
Facial muscles
|
0.947
|
0.893
|
|
Swallowing
|
0.710
|
0.815
|
|
Speech following counting aloud from 1–50
|
1.000
|
0.930
|
|
Right arm outstretched
|
0.862
|
0.761
|
|
Left arm outstretched
|
0.841
|
0.792
|
|
Forced vital capacity
|
1.000
|
0.762
|
|
Right hand grip
|
1.000
|
0.826
|
|
Left hand grip
|
0.939
|
0.876
|
|
Head lifted
|
0.772
|
0.854
|
|
Right leg outstretched
|
0.875
|
0.875
|
|
Left leg outstretched
|
0.890
|
0.890
|
Table 3
Analysis of the reproducibility by means of the Pearson’s correlation coefficient
and of the intra-class correlation coefficient values for the total score.
|
Coefficient
|
Intra-interviewer
|
Inter-interviewer
|
mean ± SD QMGS-A
|
mean ± SD QMGS-B
|
mean ± SD QMGS-A2
|
|
Pearson’s
|
0.998*
|
0.991*
|
11.43 ± 5.7
|
11.50 ± 5.2
|
11.37 ± 5.6
|
|
Intra-class
|
0.999**
|
0.994***
|
*p < 0.001; **CI = 95% (0.998–1.000); ***CI: 95% (0.987–0.997). SD: Standard Deviation;
QMGS-A: Quantitative Myasthenia Gravis score interviewer A; QMGS-B: Quantitative Myasthenia
Gravis interviewer B; Quantitative Myasthenia Gravis score interviewer A2 (second
interview).
Figure 3 Correlation between the QMGS and MGC and between the QMGS and MG-QOL15.
DISCUSSION
The use of common assessment tools is advocated by the Task Force of the Medical Scientific
Advisory Council, with the MGFA proposing a series of classification systems and definitions
of response to therapy for MG patients[7], to achieve greater uniformity in the recording and reporting of clinical trials
for MG. The QMGS is among the recommended assessment tools. However, to be applicable
to different ethnic groups, patient-report assessments, such as the QMGS, require
translation and validation for each population. A rigorous process of translation
and cultural adaptation is required to ensure congruity with the original version
of the measurement tool, facilitate uptake of the translated version, and eliminate
the risk of systematic bias of results. For these reasons, we performed the translation
and cultural adaption of the original English version of the QMGS for use with patients
with MG of Portuguese ethnicity in Brazil according to published international standards[15],[16].
The excellent intra- and inter-observer reliability and clinically acceptable concurrent
validity of the QMGS that we report in our study are comparable to values previously
reported. Sharshar et al.[12] evaluated the inter-observer reliability and concurrent validity of the QMGS against
the Myasthenia Muscle Score. Among a group of 22 patients with MG, including 13 men
and nine women, with a mean age of 63 years (range, 25 to 80 years), they reported
high inter-observer agreement and a high correlation in score between the QMGS and
Myasthenia Muscle Score. Therefore, we propose that our translated version of the
QMGS has the reliability required for its use in clinical trials and in practice.
The longitudinal construct validity of the QMGS was confirmed by Bedlack et al. in
a prospective study with 53 patients with MG[13]. In this study, the authors demonstrated the correlation between the QMGS and manual
muscle testing, as well as providing evidence of the responsiveness of the QMGS to
differentiate change in disease severity between two visits as: unchanged, improved,
or worse.
In our study, we found a very strong correlation between the QMGS and the MGC, as
well as between the QMGS and the 15-item MG-QOL. Our results are comparable to those
reported by Barnett et al.[24] who observed a good association between the 15-item MG-QOL and the QMGS. We also
confirmed the concordance between the QMGS and the MGFA classification of disease
severity, Class I to III. It is important to acknowledge that our study group did
not include patients in the MGFA Class IV and V. Barnett et al.[25] similarly confirmed the sensitivity of the QMGS to differentiate disease severity
among patients with MG, as well as to quantify treatment effectiveness. In particular,
Barnett et al. provided evidence of the discriminative value of the timed items of
the QMGS, evaluating upper and lower limb function and neck endurance.
In the process of validating the Brazilian Portuguese version of the QMGS, patients
in our study group reported excellent understanding and applicability of the QMGS,
with health professionals confirming the clinical applicability of the translated
QMGS. Moreover, reliability measures (inter- and intra-observer reliability) for the
translated QMGS were excellent, meeting the international standards for the clinical
and research use of a translated and culturally-adapted measurement tool.
In conclusion, based on our results, we can affirm that the Brazilian Portuguese translation,
cultural adaptation, and validation of the QMGS was successfully performed. We believe
that the use of our Brazilian Portuguese version of the QMGS will improve care of
patients with MG, including more precise monitoring of the clinical rehabilitation
process, as well as a more effective comparison of clinical status.
