Sage Journals: Discover world-class research

Abstract

Background:

The advent of three effective disease modifying therapies for SMA has highlighted the need to understand the epidemiology of spinal muscular atrophy (SMA) and its disability impact.

Objective:

We aimed to establish the nationwide incidence and prevalence of SMA in Aotearoa-New Zealand, and to estimate the patients’ disability and the impact of this on health resource utilisation.

Methods:

We used multiple sources to identify patients with SMA and verified the diagnosis, disabilities and resources utilisation by review of the individual patient notes and genetic results. The four year incidence period was from 1st July 2015 to 30th June 2019. Prevalence date was 1st March 2019. Of note, this time period pre-dated access to disease modifying therapy in New Zealand. Census data for 2018 was used for denominators. Descriptive statistics and capture-recapture were used to analyse the data. For context, we reviewed international SMA epidemiology.

Results:

The incidence per 100,000 live births was 8.0 (95% confidence interval (CI): 4.8–12.5). The standardised prevalence rate of SMA on 1st March 2019 was 1.78 per 100,000 (95% CI: 1.24, 2.33). Prevalence was significantly lower amongst Māori at 0.34 (95% CI: 0.08, 1.13; p = 0.006). Substantial decline from best motor milestone performance was seen; seven patients with SMA1 died without access to disease modifying therapy. 74% of the total cohort used wheelchairs. 23% required respiratory support. 62% had scoliosis, of whom 61% had had surgery. Surviving SMA1 patients had very high health service utilisation.

Conclusions:

Incidence and prevalence figures match closely with international studies. This is the first record of low SMA rates in Māori. While the largest burden of disease falls on patients with SMA1 and 2 there is still substantial use of health resources among SMA3 and SMA4 patients.

Keywords

muscular atrophy spinal epidemiology incidence prevalence ethnicity New Zealand

Introduction

We set out to investigate the epidemiology of spinal muscular atrophy (SMA) in the whole country of Aotearoa-New Zealand (population 2018, 5.02 million people). The advent of three effective disease modifying therapies for SMA (nusinersen,¹ risdiplam² and onasemnogene abeparvovec³) has made epidemiological research imperative for treatment planning. Spinal muscular atrophy (SMA) refers to a group of autosomal recessive neuromuscular diseases which are characterised by progressive degeneration of alpha motor neurones in the brainstem and spinal cord.^4,5 Most cases are caused by homozygous deletion in the survival motor neurone 1 gene (SMN1) on chromosome 5q though some patients are compound heterozygous for deletion and a pathogenic variant.⁶ Affected patients are largely dependent on their neighbouring SMN2 gene(s) for SMN protein production. As individuals have different numbers of SMN2 genes the disease severity differs with those having a greater number of SMN2 gene copies having milder disease.⁷ This leads to a classification of SMA subtypes 0–4 based on the age of onset and the best motor milestone reached⁸ (Table 1). At the time of the study none of these treatments were government funded in New Zealand so in some cases families moved overseas to access funded treatment. We studied the distribution of SMA subtypes, the age of patients, their disability and use of services, their geographical distribution and their ethnicity. While the incidence of SMA type 1 is higher than the other subtypes, SMA1 is less prevalent because of its short life expectancy without treatment. Therefore, we undertook both an incidence and a prevalence study.

Table 1.

SMA type definitions.

SMA type	Description
0	Patients have prenatal onset of disease and survive a few weeks to months. No motor milestones are achieved.
1	Patients never acquire the ability to sit independently (onset; 0–6 months).
2	Patients sit independently (onset; 6–18 month)
3	Patients are able to stand and walk without support (onset; >18-months)
3a	-age at diagnosis <3 years
3b	-age at diagnosis >3 years
4	Patients are able to walk during adulthood (onset >21 years)

Classification of SMA types is based on age at onset and maximum motor function.

An important aspect of the condition, especially when considering the cost of expensive disease modifying medical therapies is the alternative cost of treating the condition symptomatically, so we also looked into use of health resources in our medically untreated population.

In studying population figures for Aotearoa-New Zealand it is important to understand the underlying population structure. 16.5% of the population identifies as Māori – the country's indigenous people (https://www.stats.govt.nz/topics/census#2018-census). A further 8.1% identify as being from Pacific countries with the largest numbers being from Fiji, Samoa, Tonga and the Cook Islands⁹ with a relatively young demographic. “Asian” people make up 15.1% and people of European descent make up 70.2%. Total percentages exceed 100 percent because of the convention in the New Zealand census – which we used for our denominator - to allow individuals to choose more than one ethnicity (Ministry of Health. 2017. HISO 10001:2017 Ethnicity Data Protocols. Wellington: Ministry of Health).

Materials and methods

Case ascertainment

Our retrospective study of SMA prevalence and incidence collected data from multiple sources: Pūnaha Io - the NZ Neuro-Genetic Registry & BioBank (formerly the NZ NMD Registry) a voluntary registry open to patients who want to take part in research;¹⁰ patient support organisations including the Muscular Dystrophy Association of NZ; Genetic Health Services NZ; the three clinical genetics laboratories in New Zealand; Ministry of Health national disability service Needs Assessment Service Coordination (NASC) database; hospital records and self/family referrals. In addition, all nine paediatric neurologists and the 63 adult neurologists, practising either in private or public during the study period, were contacted directly regarding known diagnoses. Responses were received from all nine paediatric and 60 adult neurologists.

Data collection started in June 2019 so we set a prevalence date of 1st March 2019 and the four-year period, 1st July 2015–30th June 2019 was selected as the incidence period as this was thought to be a long enough period to accommodate natural year-to-year variation in incidence.

Disability and health resource utilisation data

Permission from the New Zealand Health and Disability Ethics Committee (#19/NTA/37) allowed public health system medical records to be obtained and reviewed without requiring patient consent, to avoid bias which might be introduced if patients could opt out. Medical records for all identified participants were requested via the Clinical Records Departments of the New Zealand public health system. In some areas these were available electronically. In other areas, and for older members of the cohort, paper files were retrieved. In New Zealand, all children with SMA are treated by public health services. Starship Children's Hospital in Auckland is a nationwide quaternary referral centre. There is no alternative private paediatric hospital service.

All medical files pertaining to the individuals’ lifespan were reviewed by one member of the research team. The following information was collected: Age, gender, ethnicity (self-identified during any health-related consultation), and geographical region at diagnosis and point prevalence date. Diagnosis and SMA subtype were confirmed by reviewing relevant medical records and genetic test results. Patients who had a positive genetic test (either homozygous 5q deletion or compound deletion and point mutation) or who had the same phenotype as a sibling with a positive genetic test were included. SMN2 copy number was collected when available, but has not been routinely tested prior to availability of disease modifying therapies. Clinical information collected from the medical record included: age at diagnosis, maximum motor milestone, current motor ability, wheelchair use, nasogastric tube or gastrostomy insertion, respiratory support (cough assist, daytime non-invasive ventilation and/or nocturnal non-invasive ventilation), scoliosis, scoliosis surgery, medications, and treatment with disease modifying therapy. The total number of inpatient hospital visits, ED visits, outpatient visits (including allied health team contacts) and operating theatre visits were counted from medical records.

Statistical analysis

Descriptive statistics were used to characterise the sample with numbers and percentages, means and standard deviations, or medians and interquartile ranges used as appropriate. Stata software (version 16) was used for all analysis.¹¹ For prevalence denominators, population estimates were taken from the census data at 30 June 2018 (https://www.stats.govt.nz/topics/census#2018-census) to allow calculations by age, sex, region and ethnic group. Following the methodology of previous studies,¹² the denominator for incidence in each 12 months of the study was the total number of live births in NZ at of the mid-point of that period (https://infoshare.stats.govt.nz/). Incidence rates were calculated overall and by study year, SMA subtype and ethnic group. The Poisson distribution was used to calculate exact 95% confidence intervals for crude prevalence and incidence rates. The overall SMA prevalence rate by ethnic group was compared using Fishers’ exact test. Age-standardisation was performed using direct standardisation to the WHO standard population 2000–2025;¹³ 95% confidence intervals for standardised rates were calculated using nonparametric bootstrap estimation with 500 replications. The Stata module RECAP was used to perform a capture-recapture analysis to estimate the number of prevalent cases missing from the data using three case ascertainment.¹¹

Literature review of international incidence and prevalence rates

We provide a survey of previous epidemiological studies of SMA with which we have compared our data, in the Supplementary Material. We also provide there the detail of how we have approached selection, inclusion and analysis of the data.

Results

Patients identified

Eighty patients with SMA were identified as either being incident in the observation period (19 patients) or living in Aotearoa-New Zealand on the study census day (71 patients). Of the 19 incident patients just ten were also counted in the prevalence cohort, nine having died or moved overseas (see also Supplementary Table 1). There were 58 singletons, ten affected siblings-pairs and two who had an affected sibling who were not included.

These 80 patients were distributed amongst SMA types as follows: SMA0 (1), SMA1 (9), SMA2 (31), SMA3 (37), SMA4 (2). Sex ratio was almost exactly half-and-half male:female Just 3.9% of the sample identified as Māori compared with 16.5% of people in the general population (Table 2). Seventy-nine of the 80 patients had homozygous SMN1 deletion with just one person found to be homozygous for a point mutation. One person's genetic diagnosis was attributed on the basis of their similarly affected sibling's positive diagnosis – a practice not-uncommon in genetic diseases in the past when costs of genetic testing were higher than more recently. There were just 9 cases with a known SMN2 copy number, of which 2 cases had 2 copies (1 with SMA2 and 1 with SMA3) and 7 cases had 3 copies (4 with SMA2, 2 with SMA3 and 1 with SMA4).

Table 2.

SMA subtypes and demographics of whole study sample.

	SMA1	SMA2	SMA3	SMA4	All SMA
	(N = 9)	(N = 31)	(N = 37)	(N = 2)	(N = 80)^a
Age at diagnosis (years)
Median (IQR)	0.8(0.2, 1.1)	1.7(1.2, 2.0)	4.0 (2.0, 14.0)	-	2.0(1.2, 6.0)
Range (youngest – oldest)	0–1	0–19	1–62	-	0–62
Sex
Female	3 (33.3%)	15 (48.4%)	21 (58.3%)	1 (50.0%)	41 (50.6%)
Male	6 (66.7%)	16 (51.6%)	15 (41.7%)	1 (50.0%)	40 (49.4%)
Ethnicity
NZ European	-	-	-	-	54 (66.7%)
Other European	-	-	-	-	6 (7.4%)
Māori	-	-	-	-	3 (3.7%)
Pacific	-	-	-	-	6 (7.4%)
Asian	-	-	-	-	10 (12.3%)
Other	-	-	-	-	2 (2.5%)

Total includes 1 SMA0 patient not otherwise detailed.

Incidence

The average annual incidence was 8.0 (4.8–12.5, 95% CI) per 100,000 live births (Table 3). The rate was reasonably consistent over the four years. Roughly a third of patients were diagnosed with each of SMA1, 2 and 3 with just one additional patient diagnosed with type 0 and no type 4. Fourteen were NZ European/Other European, 1 Māori, 2 Pacific People, 1 Asian and 1 “Other”. Incidence rates per 100,000 live births (95% CI) by ethnicity were: NZ European/Other European/Other 13.5 (7.6, 22.3); Māori 1.7 (0.0, 9.3); Pacific People 8.3 (1.0, 30.0) and Asian 2.4 (0.0, 13.2) (see Supplementary Table 3).

Table 3.

SMA incidence in NZ, comprising 19 cases diagnosed in NZ between 1st July 2015 and 30th June 2019.

Time period	Incident cases in NZ	Live births in NZ^a	Incidence rate, per 100,000 live births (95% CI)^b
Overall
July 2015–June 2019	19	238,098	8.0 (4.8, 12.5)
Annually
July 2015–June 2016	4	61,038	6.6 (1.8, 16.8)
July 2016–June 2017	4	59,430	6.7 (1.8, 17.2)
July 2017–June 2018	7	59,610	11.7 (4.7, 24.2)
July 2018–June 2019	4	58,020	6.9 (1.9, 17.7)
By subtype
SMA0	1	238,098	0.4 (0.00, 2.3)
SMA1	6	238,098	2.5 (0.9, 5.5)
SMA2	6	238,098	2.5 (0.9, 5.5)
SMA3	6	238,098	2.5 (0.9, 5.5)
SMA4	0	238,098	-
SMA0/SMA1	7	238,098	2.9 (1.2, 6.1)
SMA2/SMA3	12	238,098	5.0 (2.6, 8.8)
By Ethnicity^c
NZ European/Other	15	111,010	13.5 (7.6, 22.3)
Māori	1	60,010	1.7 (0.0, 9.3)
Pacific People	2	24,113	8.3 (1.0, 30.0)
Asian	1	42,172	2.4 (0.0, 13.2)

Denominator for incidence in each time period is the number of live births in NZ as of the mid-point of that period, i.e. in December (Stats NZ).

95% CIs calculated using the Poisson distribution.

Birthrate estimates by ethnicity for the period July 2015 – June 2019 were calculated using full calendar year data, with 2015 and 2019 esimtates halved. Data from the Ministry of Health «Report on Maternity» web tool, which includes only live births recorded in the Maternity Collection. https://minhealthnz.shinyapps.io/report-on-maternity-web-tool/.

Prevalence

The overall standardised prevalence rate of SMA in Aotearoa-New Zealand on 1st March 2019 was 1.78 per 100,000 (95% CI: (1.24, 2.33)) (Table 4). Of the 19 cases in the incidence analysis, seven (one with SMA0, six with SMA1) had died before the prevalence date. Five were < 6 months old at time of death, and two between 6 months and 3 years old). A further two patients (one SMA2 one SMA3) had moved overseas (to access disease modifying therapy). The overall median age for all 71 patients was 19.0 years (interquartile range 8.0,28.0). The median age for each of the subtypes were 6.0 years (5.0, 8.0) for SMA1; 14.0 years (6.0,25,0) for SMA2; 21.0 years (15.0, 36.0) for SMA3 and 27.5 years for SMA4.

Table 4.

SMA prevalence for n = 71 cases alive and living in NZ on 1^st March 2019.

Grouping	SMA cases in NZ 1^st March 2019		Crude prevalence per 100,000 (95% CI)^a,b	Standardised prevalence per 100,000 (95% CI)^a,c
Grouping	N	%	Crude prevalence per 100,000 (95% CI)^a,b	Standardised prevalence per 100,000 (95% CI)^a,c
Total	71	100.0%	1.49 (0.73, 2.25)	1.78 (1.24, 2.33)
SMA type
1	3	4.2%	0.06 (0.01, 0.19)	0.08 (0.00, 0.20)
2	30	42.3%	0.64 (0.43, 0.91)	0.79 (0.46, 1.13)
3	36	50.7%	0.77 (0.54, 1.06)	0.88 (0.61, 1.15)
4	2	2.8%	0.04 (0.01, 0.15)	0.05 (0.00, 0.09)
Ethnicity
NZ European/Other European/Other	54	76.0%	1.58 (1.18, 2.06)	1.94 (1.31, 2.57)
Māori	3	4.2%	0.39 (0.08, 1.13)	0.34 (0.05, 0.63)
Pacific People	5	7.0%	1.31 (0.43, 3.06)	1.17 (0.55, 1.79)
Asian	9	12.7%	1.27 (0.58, 2.41)	1.35 (0.39, 2.30)

Population denominator estimate 4,699,755, taken from the census data at 30 June 2018 (Stats NZ).

95% CIs calculated using the Poisson distribution for crude rates.

Age-standardised using direct standardisation to the WHO standard population 2000–2025; 95% CIs calculated using nonparametric bootstrap estimation with 500 replications for standardised rates.

As suggested by the incidence results, a substantial and statistically significantly (Fishers’ exact test p = 0.006) lower prevalence rate was also seen amongst Māori – about one-sixth that seen in the population overall (Table 4). Prevalence rates in Pacific people and Asians were slightly lower than that for Europeans/others; however, these differences were not statistically significant (p = 0.83 and p = 0.09, respectively), and estimates for these three ethnic groups were broadly consistent with the average rate for the country (Figure 1).

We explored the possibility that patients in more remote settings might have a lower likelihood of diagnosis. We found no evidence to support this with just one region (Wellington - an urban centre) where the 95% CI did not cross the overall average (see Supplementary Figure 1). The result for Wellington probably just reflects random variation in a small sample.

Capture-recapture analysis

We received data from four different data sources (Supplementary Table 3 and Supplementary Figure 2). The vast majority of cases (97.2%) were identified from doctor records. The Needs Assessments Agencies identified only two cases, which were both also identified in two or three other sources. Cases identified via Pūnaha Io – the NZ NeuroGenetic Registry and Biobank were all also identified through another source. All but three cases had at least two sources of ascertainment. In two of these cases the only source of identification was via the lab, and in the third, the only source was doctor records. A capture-recapture analysis in Stata using the 3 main lists (doctors, lab and NMD Registry – total N = 71) gave an estimated N of 71 (95% CI 71,74).¹¹

Capture-recapture analysis relies on i) different sources identifying the same person in a way that enables us to know they’re the same person – for our dataset all sources use the same National Health Index number ii) accounting for transient patients might appear in one dataset but not another, as the point prevalence date was the same for all databases this did not occur; iii) Patients have an equal chance of being identified by one of the methods. This very high response rate from New Zealand neurologists, the national coverage of Pūnaha Io and the fact that all patients diagnosed in New Zealand would have had their samples assessed in one of the three laboratories satisfies this criterion. Patients diagnosed outside New Zealand who never consulted a neurologist would not have been picked up this way but these are likely to be very few.

Disability and health resource utilisation

The data from the prevalence sample are presented here. Only three SMA1 patients were alive in our cohort at the point prevalence date with a median age of 6 years (see Table 5). All definitely satisfied the diagnostic criteria for SMA1 in that they had symptom onset before six months of age, and never learned to sit independently. Amongst the SMA2 patients, a range of best motor milestones was seen. All had achieved independent sitting. The median maximum motor milestone achieved was only just greater than that at crawling; the upper quartile range was standing with support. All type 3 and 4 patients had achieved independent walking but there had been substantial decline amongst the SMA3 group with almost half requiring support even to sit by prevalence day. Of the patients with SMA3a (onset before age 3), 68% were permanent wheelchair users and 14% were intermittent wheelchair users. Amongst patients with SMA3b the respective figures were 8% and 17%. Both SMA4 patients were still ambulant (See Supplementary Table 2).

Table 5.

Maximum and prevalence day motor milestones, SMA complications and interventional management in prevalence sample.

	SMA1	SMA2	SMA3	SMA4	All SMA
	(N = 3)	(N = 30)	(N = 36)	(N = 2)	(N = 71)
Age at prevalence day	6.0	14.0	21.0	27.5	19.0
Years: median (IQR)	^a	(6.0, 25.0)	(15.0, 36.0)	^a	(8.0, 29.0)
Maximum motor milestone
Head control	-	-	-	-	-
Sitting with support	3 (100.0%)	-	-	-	3 (4.3%)
Independent sitting	-	12 (41.4%)	-	-	12 (17.1%)
Crawling	-	5 (17.2%)	-	-	5 (7.1%)
Standing with support	-	6 (20.7%)	-	-	6 (8.6%)
Standing	-	1 (3.4%)	-	-	1 (1.4%)
Assisted walking	-	5 (17.2%)	-	-	5 (7.1%
Independent walking	-	-	36 (100.0%)	2 (100.0%)	38 (54.3%)
Motor ability (prevalence day)
Head control	-	1 (3.4%)	-	-	1 (1.4%)
Sitting with support	3 (100.0%)	21 (72.4%)	15 (41.2%)	-	39 (54.9%)
Independent sitting	-	2 (6.9%)	2 (5.6%)	-	4 (5.7%)
Crawling	-	2 (6.9%)	-	-	2 (2.9%)
Standing with support	-	-	-	-	-
Standing	-	-	1 (2.8%)	-	1 (1.4%)
Assisted walking	-	3 (10.3%)	1 (2.8%)	-	4 (5.7%)
Independent walking	-	-	17 (47.2%)	2 (100.0%)	19 (26.8%)
Wheelchair use
None	-	3 (10.0%)	12 (34.3%)	2 (100.0%)	18 (25.7%)
Intermittent use	-	4 (13.3%)	6 (17.1%)	-	10 (14.3%)
Permanent	3 (100.0%)	23 (76.7%)	17 (48.6%)	-	42 (60.0%)
Respiratory support
Any	3 (100%)	19 (63.3%)	3 (8.3%)	0 (0.0%)	21 (29.6%)
Cough assist only	0 (0.0%)	3 (10.0%)	2 (5.6%)	0 (0.0%)	5 (7.0%)
NIV only	1 (33.3%)	4 (40.0%)	1 (2.8%)	0 (0.0%)	6 (8.4%)
Both Cough assist and NIV	2 (66.7%)	8 (30.0%)	0 (0.0%)	0 (0.0%)	10 (12.7%)
Total, Cough assist	2 (66.7%)	11 (36.7%)	2 (5.7%)	0 (0.0%)	15 (21.1%)
Total, NIV	3 (100%)	12 (40.0%)	1 (2.9%)	0 (0.0%)	16 (22.5%)
Mean age at first NIV use	0.83^a	6.0 (3.0, 12.0)	15^a	-	5.5 (2.0, 12.0)
[n with data]	²	¹¹	¹		¹⁴
Feeding Support
Nasogastric tube	0 (0.0%)	1 (3.3%)	0 (0.0%)	0 (0.0%)	1 (1.4%)
Gastrostomy	2 (66.7%)	11 (36.7%)	0 (0.0%)	0 (0.0%)	13 (18.6%)
Scoliosis
Diagnosed	3 (100.0%)	26 (86.7%)	15 (42.9%)	0 (0.0%)	44 (62.9%)
Surgery	1 (33.3%)	18 (60.0%)	8 (22.9%)	0 (0.0%)	27 (38.6%)
Mean age at scoliosis surgery	5^a	8.0 (7.0, 11.0)	11.5 (10.5, 14.5	-	9.0 (7.0, 12.0)
[n with data]	¹	¹¹	⁸		¹⁵

Where n is too small IQR is omitted. IQR, interquartile range; NIV, non-invasive positive pressure ventilation. Observations up until Prevalence Day are included. Date for first use of wheelchair and first use of feeding support not available.

Almost a third of all patients used either cough assist devices, non-invasive ventilation (NIV) or both (Table 5), including all patients with SMA1 and 40% of those with SMA2. Positive pressure ventilation was seldom required in SMA3 (n = 1 patient with SMA3a), and not used at all in SMA4. Tube feeding was required for 40% of patients with SMA2, almost all of whom had a gastrostomy inserted for this purpose. No patients with SMA3 or 4 required tube feeds. 62% of all SMA patients developed a scoliosis, including all patients with SMA1, 87% of patients with SMA2, and 43% of patients with SMA3 (including two SMA3b patients). Over half the patients with scoliosis (61%) required spinal surgery. For those for whom data were available, the median age of first NIV use was 5.5 (2.0, 12.0) years and scoliosis surgery 9.0 (7.0,12.0) years (Table 5).

Utilisation of healthcare services was high across all SMA subtypes. The three SMA1 patients had had a median 1045 out-patient appointments (medical and allied health), 22 inpatient admissions and 6 visits to theatre each (Table 6). SMA2 patients also had very high healthcare demands with the 31 patients undergoing a median 89 out-patient visits, 7 inpatient visits, and 2 theatre visits. The 37 SMA3 patients still required significant resources with a median 33 outpatient visits each as well as one in-patient admission and one theatre visit.

Table 6.

Lifetime per-patient reported hospital visits for prevalence sample.

		SMA1	SMA2	SMA3	All SMA^a
DHB	Type of visit	(N = 3)	(N = 30)	(N = 36)	(N = 71)
		Median Age 6.0	Median Age 14.0	Median Age 21.0	Median Age 27.5
Auckland Metro DHBs	Inpatient	22 (22, 49)	12 (4, 21)	3 (0, 5)	5 (2, 18)
	Outpatient	1045 (114, 1711)	41 (12, 132)	52 (12, 133)	56 (12, 136)
	Theatre	6 (6, 9)	2 (1, 3)	1 (0, 2)	1 (1, 3.5)
Other DHBs	Inpatient	3 (3, 3)	10 (5, 16)	1 (1, 3.5)	3 (1, 10)
	Outpatient	1 (1, 1)	71 (23, 221)	26 (10, 52)	30 (15, 109)
	Theatre	-	3 (1, 5)	1 (1, 2)	2 (1, 3)
Total	Inpatient	25 (25, 25)	23 (12, 38)	4 (3, 20)	19 (6, 32)
	Outpatient	1045 (115, 1711)	122 (48, 221)	44 (20, 94)	89 (27, 161)
	Theatre	6 (6, 9)	2 (1, 5)	1 (0, 2)	2 (1, 5)

Data presented are the median (interquartile range) of the total reported hospital visits over an individual's life (up to the point prevalence date). DHB, District Health Board. The Auckland Metro DHBs are grouped together as Paediatric and Adult Neurology services in this are regionwide. The vast majority of SMA1 patient interactions occur in this section of the data because the Paediatric hospital is a nationwide referral centre for severely ill children. Age is median age in years for each subtype. Outpatient includes appointments with Allied Health and medical specialists.

Incomplete information available for one SMA4 patient meant that we could only report one patient's data.

Discussion

The advent of three effective disease modifying therapies for SMA has highlighted the need to understand the epidemiology of SMA for and its disability impact to inform healthcare planning. We present here the results of a thorough retrospective, four-year incidence, and point prevalence study of SMA in the whole population of Aotearoa-New Zealand. The incidence of SMA was 8.0 cases per 100,000 live births (95% CI: (4.8, 12.5)), which places Aotearoa-New Zealand in the middle of international estimates of prevalence. SMA subtype incident rates were also in keeping with the literature. The age standardised prevalence on 1^st March 2019 was 1.78 per 100,000 (95% CI: (1.24, 2.33)) – near the high end of the range of previous studies. However, other studies have not provided age standardised results which makes it difficult to make full comparisons. Age-standardised prevalence figures for SMA subtypes were also generally consistent with previously observed rates though our study found a higher SMA2 prevalence rate than the two previous studies. We are the first study to give a prevalence estimate for SMA4.

Capture-recapture analysis suggests that all patients with a diagnosis of SMA in Aotearoa-New Zealand were found. While it is possible that not all patients with SMA are diagnosed, infants with SMA1 are likely to present to paediatricians and neurologists because of the severity of their disease and genetic testing is widely available and publicly funded so it is likely that most patients with SMA1 have been found in our study. However, because genetic testing for SMA has only been readily available in Aotearoa-New Zealand since 2007 it is possible that older patients with SMA3 and SMA4 have not been diagnosed. This would be consistent with the upper interquartile range for SMA3 being just 36 years. With the introduction of disease modifying therapies more patients with adult SMA may present for diagnosis.

Almost all incidence and prevalence rates (totals and SMA subtypes) from our study lie within the range of those previously observed (Table 7). The only exception was our estimate for the standardised prevalence of SMA2, (0.79/100,000, 95% CI 0.46–1.13) which was higher than the two previous estimates (0.13 and 0.53). It is also interesting to note the total SMA rates are within the range of those reported in large newborn screening programmes such as in a Russian study 12.8/100,000 (26 out of 202 908 live births¹⁶ and national statistics in the USA of 6.8/100,000 (425 out of 6 244 825) live births.¹⁷

Table 7.

Present study incidence and prevalence compared with literature.

	Incidence / 100,000 live births		Prevalence /100,000
SMA type	Present study	Previous studies	Present study	Previous studies
SMA type	SI (95%CI)	Average (range) : N	SPP (95% CI)	Average (range) : N
All SMA	8.0 (4.8–12.5)	8.8 (4.4–13.7) : 13	1.78 (1.24–2.33)	1.14 (0.56–1.9) : 4
SMA1	2.9 (1.2–6.1)	5.3 (1.7–9.8) : 13	0.08 (0.00–0.20)	0.13 (0.04–0.24) : 7
SMA2	2.5 (0.9–5.5)	2.7 (1.0–5.3) : 8	0.79 (0.46–1.13)	0.33 (0.13–0.53) : 2
SMA3	2.5 (0.9–5.5)	2.1 (0.47–4.6) : 8	0.88 (0.61–1.15)	0.67 (0.13–1.2) : 2
SMA4	-	-	0.05 (0.00–0.09)	-
SMA2 and SMA3	5.0 (2.6–8.8)	5.0 (1.6–10.1) : 11	1.68 (1.17–2.18)	1.73 (0.26–2.5) : 5

SI, Standardised Incidence; SPP, Standardised point prevalence; CI, confidence interval (see Supplementary Tables 5 and 6) – note the figures above are based on the following papers.^14,18–46

Of particular note, our “All SMA” prevalence estimate of 1.78/100,000 was close to the most similarly conducted prevalence study by Norwood et al.¹⁴ which showed a prevalence of 1.9/100,000. Generally, our SMA1 incidence and prevalence were at the lower level of that reported whereas our combined figures for SMA2 and 3 almost exactly mirror the literature. We are the first to report a prevalence for SMA4 and the only study to report population standardised data.

This study was performed at a time when New Zealand patients did not have access to disease modifying therapies. It was notable that many of the patients identified in the incidence part of the study were no longer living in Aotearoa-New Zealand on prevalence day. Seven patients with SMA0 or 1 died without access to therapy, and two patients had moved overseas in the hope of receiving disease modifying therapy. Both of these outcomes are devastating and potentially preventable with funded access to powerful RNA based therapies such as nusinersen and risdiplam and also gene therapy with onasemnogene abeparvovec. Furthermore, it gives some context to the surprising survival of three infants with SMA1 beyond the natural history expected for infants with SMA1 – these are the minority - prevalence studies naturally select patients with longer survival.

While no study has found a population free of SMA, disease prevalence and carrier frequencies have been shown to differ across ethnic groups.^15,18 The substantially lower observed prevalence amongst New Zealand Māori where an overall standardised prevalence of 0.34 cases per 100,000 people was seen may be another example of this. Alternatively, it could represent under-reporting due to the known barriers to healthcare for Māori.^47,48 Our incidence figures were consistent, and the finding of higher rates amongst Pacific People (almost three times higher) and rates in rural areas similar to urban (Supplementary Figure 1 footnotes) – both populations known also to experience barriers to healthcare in Aotearoa-New Zealand^49,50 suggest that this is likely to be a true finding. This is important within Aotearoa-New Zealand from points of view of equity for our indigenous people and for future projections of healthcare needs as the proportion of Māori in the population increase (Stats NZ).

Another important finding of our study is that one in six families had more than one affected sibling. While government-funded genetic counselling is available to all families affected by a diagnosis of SMA and two cycles of IVF/preimplantation genetic diagnosis have been funded since 2005, in the vast majority of families, the couple was already pregnant or the second child born by the time their older child was diagnosed. The potential to avoid a second affected pregnancy highlights the benefit of new-born or preconception screening.⁵¹

This study quantifies the very significant direct health impacts of untreated SMA on the medical system in New Zealand. Patients with SMA1 had especially high numbers of outpatient appointments. Healthcare encounter rates were progressively lower in SMA2 and SMA3 but given the overall numbers of prevalent patients, still represent a significant health resource and individual burden. The need for support with ambulation/mobility, ventilation and scoliosis management was very high across all patients. Three quarters required assistance with ambulation most commonly requiring a wheelchair. Almost 30% required ventilatory support and almost 20% ongoing tube feeding. Scoliosis affected almost two thirds and close to 40% underwent scoliosis surgery. Standards of Care guidelines are vital in ensuring patients have active surveillance and management of these complications. This becomes increasingly important, alongside disease modifying therapy, to maximise the benefits of treatment as we seek to best understand the new phenotype of treated SMA. Disease modifying therapies offer the best outcomes with early, ideally presymptomatic access to treatment in order to prevent the irreversible loss of motor neurones.^52,53 Increasing health economic analyses support newborn screening and pre-symptomatic treatment as the best way to reduce morbidity and health care burden.^15,54,55

The main potential weakness of the study is its retrospective nature. Determination of current clinical status and motor milestone achievement for each patient was limited by the frequency at which medical review and milestone assessments were performed. Although this is the one of the largest studies to date where clinical diagnosis (not just genetic status) was individually verified, the total numbers are relatively small. However, these issues are offset by a number of features. In particular, we calculated the prevalence on a whole population basis and aimed to identify not just children but adults with SMA. The fact that our study found patients through multiple sources allowed for a capture-recapture analysis which gave reassuring results as to the comprehensiveness of the study. It is the first study to calculate standardised prevalence rates, which will allow for more accurate international comparisons with future studies.

In conclusion, overall the prevalence of SMA in New Zealand is similar to international figures. The substantially lower prevalence and incidence seen in Māori in this study warrants further investigation and will be one of the outcomes that may be assessed when newborn screening is introduced. The study also underscores the high rates of morbidity and heavy use of health resources associated with untreated SMA - highest amongst patients with SMA types 1 and 2, but still substantial amongst types 3 and 4. Going forward, the quantification of morbidity will provide a baseline for evaluating of the impact of disease modifying therapy when used to treat symptomatic or pre-symptomatic infants.

Supplemental Material

sj-pdf-1-jnd-10.1177_22143602251319165 - Supplemental material for Epidemiology of spinal muscular atrophy in Aotearoa-New Zealand

Supplemental material, sj-pdf-1-jnd-10.1177_22143602251319165 for Epidemiology of spinal muscular atrophy in Aotearoa-New Zealand by Richard H Roxburgh, Alana Cavadino, Miriam Rodrigues, Sharron Meadows, Juliette Meyer and Gina O'Grady in Journal of Neuromuscular Diseases

Footnotes

Acknowledgements

The original literature search and strategy was devised by Prof Alice Theadom and we learned much from her approach to neuroepidemiology.

ORCID iD

Richard H. Roxburgh

Funding

The study was funded by Biogen (Australia New Zealand) as an Investigator initiated project.

Competing interests

RR, MR and GO’G have served on an NZ Advisory board for Risdiplam (Roche).

MR and GO'G have served on an NZ Advisory board for nusinersen (Biogen).

Data availability statement

The data supporting the findings of this study are available within the article and/or its .

Supplemental material

Supplemental material for this article is available online.

References

Chiriboga

Swoboda

Darras

, et al. Results from a phase 1 study of nusinersen (ISIS-SMN(Rx)) in children with spinal muscular atrophy. Neurology 2016; 86: 890–897.

Dhillon

. Risdiplam: first approval. Drugs 2020; 80: 1853–1858.

Dabbous

Maru

Jansen

, et al. Survival, motor function, and motor milestones: comparison of AVXS-101 relative to nusinersen for the treatment of infants with spinal muscular atrophy type 1. Adv Ther 2019; 36: 1164–1176.

Werdnig

. Zwei fruhinfantile hereditre Falle von progressiver Muskelatrophie unter dem Bilde der Dystrophie, aber auf neurotischer Grundlage. Arch Psychiat Nervenkrank 1891; 22: 437–480.

Hoffman

. Über chronische spinale Msukelatrophie im Kindesalter, auf familiärer Basis. Deutsche Zeitschrift fur Nervenheilkunde 1893; 3: 427–470.

Lefebvre

Bürglen

Reboullet

, et al. Identification and characterization of a spinal muscular atrophy-determining gene. Cell 1995; 80: 155–165.

Gavrilov

Shi

Das

, et al. Differential SMN2 expression associated with SMA severity. Nat Genet 1998; 20: 230–231.

Finkel

Bertini

Muntoni

, et al. 209th ENMC international workshop: outcome measures and clinical trial readiness in spinal muscular atrophy 7-9 November 2014, Heemskerk, The Netherlands. Neuromuscul Disord 2015; 25: 593–602.

Statistics New Zealand and Ministry of Pacific Island Affairs. Demographics of New Zealand's Pacific population. Wellington: Statistics New Zealand, Tautauranga Aotearoa, 2010.

10.

Rodrigues

O'Grady

Hammond-Tooke

, et al. The New Zealand neuromuscular disease patient registry; five years and a thousand patients. J Neuromuscul Dis 2017; 4: 183–188.

11.

der Heiden

. RECAP: Stata module to perform capture-recapture analysis for three sources with Goodness-of-Fit based confidence intervals. 2009.

12.

Verhaart

IEC

Robertson

Wilson

, et al. Prevalence, incidence and carrier frequency of 5q-linked spinal muscular atrophy - a literature review. Orphanet J Rare Dis 2017; 12: 124.

13.

Ahmed

Paterson

Warren

, et al. Biomarkers in dementia: clinical utility and new directions. J Neurol Neurosurg Psychiatry 2014; 85: 1426–1434.

14.

Norwood

Harling

Chinnery

, et al. Prevalence of genetic muscle disease in Northern England: in-depth analysis of a muscle clinic population. Brain 2009; 132: 3175–3186.

15.

Dangouloff

Hiligsmann

Deconinck

, et al. Financial cost and quality of life of patients with spinal muscular atrophy identified by symptoms or newborn screening. Dev Med Child Neurol 2023; 65: 67–77.

16.

Efimova

Zinchenko

Marakhonov

, et al. Epidemiology of spinal muscular atrophy based on the results of a large-scale pilot project on 202,908 newborns. Pediatr Neurol 2024; 156: 147–154.

17.

Belter

Taylor

Jorgensen

, et al. Newborn screening and birth prevalence for spinal muscular atrophy in the US. JAMA Pediatr 2024; 178: 946–949.

18.

Winsor

Murphy

Thompson

, et al. Genetics of childhood spinal muscular atrophy. J Med Genet 1971; 8: 143–148.

19.

Zellweger

Hanhart

. The infantile proximal spinal muscular atrophies in Switzerland. Helv Paediatr Acta 1972; 27: 355–360.

20.

Pearn

. The gene frequency of acute Werdnig-Hoffmann disease (SMA type 1). A total population survey in North-East England. J Med Genet 1973; 10: 260–265.

21.

Pearn

. Incidence, prevalence, and gene frequency studies of chronic childhood spinal muscular atrophy. J Med Genet 1978; 15: 409–413.

22.

Radhakrishnan

el-Mangoush

Gerryo

. Descriptive epidemiology of selected neuromuscular disorders in Benghazi, Libya. Acta Neurol Scand 1987; 75: 95–100.

23.

Czeizel

Hamula

. A Hungarian study on Werdnig-Hoffmann disease. J Med Genet 1989; 26: 761–763.

24.

Ignatius

Donner

. Epidemiology of childhood spinal muscular atrophy (in Finnish). In: Kiasma

(ed) Lihastautien Kehittyvae Tutkimus Ja Hoito. Turku, 1989.

25.

Spiegler

Hausmanowa-Pertrusewicz

Borkowska

, et al. Population data on acute infantile and chronic childhood spinal muscular atrophy in Warsaw. Hum Genet 1990; 85: 211–214.

26.

MacMillan

Harper

. Single-gene neurological disorders in South Wales: an epidemiological study. Ann Neurol 1991; 30: 411–414.

27.

Burd

Short

Martsolf

, et al. Prevalence of type I spinal muscular atrophy in North Dakota. Am J Med Genet 1991; 41: 212–215.

28.

Merlini

Stagni

Marri

, et al. Epidemiology of neuromuscular disorders in the under-20 population in Bologna province, Italy. Neuromuscul Disord 1992; 2: 197–200.

29.

Mostacciuolo

Danieli

Trevisan

, et al. Epidemiology of spinal muscular atrophies in a sample of the Italian population. Neuroepidemiology 1992; 11: 34–38.

30.

Thieme

Mitulla

Schulze

, et al. Epidemiological data on Werdnig-Hoffmann disease in Germany (West-Thüringen). Hum Genet 1993; 91: 295–297.

31.

Thieme

Mitulla

Schulze

, et al. Chronic childhood spinal muscular atrophy in Germany (West-Thüringen)–an epidemiological study. Hum Genet 1994; 93: 344–346.

32.

Ludvigsson

Olafsson

Hauser

. Spinal muscular atrophy. Incidence in Iceland. Neuroepidemiology 1999; 18: 265–269.

33.

Darin

Tulinius

. Neuromuscular disorders in childhood: a descriptive epidemiological study from western Sweden. Neuromuscul Disord 2000; 10: 1–9.

34.

Vaidla

Talvik

Kulla

, et al. Descriptive epidemiology of spinal muscular atrophy type I in Estonia. Neuroepidemiology 2006; 27: 164–168.

35.

Zaldívar

Montejo

Acevedo

, et al. Evidence of reduced frequency of spinal muscular atrophy type I in the Cuban population. Neurology 2005; 65: 636–638.

36.

Arkblad

Tulinius

Kroksmark

, et al. A population-based study of genotypic and phenotypic variability in children with spinal muscular atrophy. Acta Paediatr 2009; 98: 865–872.

37.

Kekou

Svingou

Sofocleous

, et al. Evaluation of genotypes and epidemiology of spinal muscular atrophy in Greece: a nationwide study spanning 24 years. J Neuromuscul Dis 2020; 7: 247–256.

38.

Kostera-Pruszczyk

Napiórkowski

Szymańska

, et al. Spinal muscular atrophy: epidemiology and health burden in children - a Polish national healthcare database perspective before introduction of SMA-specific treatment. Neurol Neurochir Pol 2021; 55: 479–484.

39.

Sarv

Kahre

Vaidla

, et al. The birth prevalence of spinal muscular atrophy: a population specific approach in Estonia. Front Genet 2021; 12: 796862.

40.

Jedrzejowska

Milewski

Zimowski

, et al.

Incidence of spinal muscular atrophy in Poland–more frequent than predicted?

Neuroepidemiology 2010; 34: 152–157.

41.

Verhaart

IEC

Robertson

Leary

, et al. A multi-source approach to determine SMA incidence and research ready population. J Neurol 2017; 264: 1465–1473.

42.

Tangsrud

Halvorsen

. Child neuromuscular disease in southern Norway. Prevalence, age and distribution of diagnosis with special reference to “non-Duchenne muscular dystrophy”. Clin Genet 1988; 34: 145–152.

43.

Radhakrishnan

Thacker

Maloo

. A clinical, epidemiological and genetic study of hereditary motor neuropathies in Benghazi, Libya. J Neurol 1988; 235: 422–424.

44.

Hughes

Hicks

Nevin

, et al. The prevalence of inherited neuromuscular disease in Northern Ireland. Neuromuscul Disord 1996; 6: 69–73.

45.

Chung

Wong

. Prevalence of neuromuscular diseases in Chinese children: a study in southern China. J Child Neurol 2003; 18: 217–219.

46.

Okamoto

Motoki

Saito

, et al. Survey of patients with spinal muscular atrophy on the island of Shikoku, Japan. Brain Dev 2020; 42: 594–602.

47.

Rumball-Smith

. Not in my hospital? Ethnic disparities in quality of hospital care in New Zealand: a narrative review of the evidence. N Z Med J 2009; 122: 68–83.

48.

Ellison-Loschmann

Pearce

. Improving access to health care among New Zealand's Maori population. Am J Public Health 2006; 96: 612–617.

49.

Ryan

Grey

Mischewski

. Tofa Saili: a review of evidence about health equity for Pacific Peoples in New Zealand. Wellington, 2019. 27/5/2023.

50.

Harwood

MLN

Ranta

Thompson

, et al. Barriers to optimal stroke service care and solutions: a qualitative study engaging people with stroke and their whānau. N Z Med J 2022; 135: 81–93.

51.

Ciarleglio

Bennett

Williamson

, et al. Genetic counseling throughout the life cycle. J Clin Invest 2003; 112: 1280–1286.

52.

Strauss

Farrar

Muntoni

, et al. Onasemnogene abeparvovec for presymptomatic infants with two copies of SMN2 at risk for spinal muscular atrophy type 1: the phase III SPR1NT trial. Nat Med 2022; 28: 1381–1389.

53.

Strauss

Farrar

Muntoni

, et al. Onasemnogene abeparvovec for presymptomatic infants with three copies of SMN2 at risk for spinal muscular atrophy: the phase III SPR1NT trial. Nat Med 2022; 28: 1390–1397.

54.

Weidlich

Servais

Kausar

, et al. Cost-effectiveness of newborn screening for spinal muscular atrophy in England. Neurol Ther 2023; 12: 1205–1220.

55.

Shih

Farrar

Wiley

, et al. Newborn screening for spinal muscular atrophy with disease-modifying therapies: a cost-effectiveness analysis. J Neurol Neurosurg Psychiatry 2021; 92: 1296–1304.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.43 MB