4 DISCUSSION
SMA of type 1 is more severe, and both the age at death and the number of people using ventilators are respectively earlier and more than those in other SMA types.1,10 Most children with SMA type 1 over two years of age need tracheotomy or all-day noninvasive ventilation for survival.11 Children with type 2 have a progressive decline in overall function and need noninvasive ventilation from five to thirteen years old.12Similarly, our study showed that the risk for RRTIs and/or ARF in type 1 is approximately four times higher than that in types 2 and 3. The average age of type 1 patients in our study was over two years, but none of them had received disease-modifying medications or long-term mechanical ventilation therapies. Among the 14 cases of type 1 with SMN2 copy number detected, five cases had 2 copies, and nine cases had 3 copies, which indicated the milder SMA types 1b and 1c.1 This was because most of the patients who could come to Beijing for treatment from other areas of China would have had mild symptoms, and the type 1 population in China might carry more copies of SMN2.13 Therefore, if all patients were considered, the risk of type 1 RRTIs and/or ARF should be higher.
Malnutrition is common in children with SMA. Our study showed that age-specific BMIz decreased in all three types and was lowest in type 1, which was consistent with other studies.14Malnutrition in SMA is always related to the severity of pulmonary complications. For example, children with dysphagia often have aspiration pneumonia and respiratory distress and rely on mechanical ventilation earlier.15 The age at gastrostomy is related to the first appearance of ARF and initiation of continuous mechanical ventilation.11 Similarly, our study showed that BMIz is an independent risk factor for RRTIs and/or ARF in children with SMA, and the risk increases by approximately one and a half times for every one unit decrease in BMIz. In terms of mechanism, masticatory muscle weakness, dysphagia and respiratory problems will reduce the intake of calories, while the work of breathing will increase energy consumption. As a result, the more severe the children are, the higher their risk of malnutrition.1 Therefore, it is crucial to pay attention to the BMIz level to prevent the occurrence of RRTIs and/or ARF. The appropriate intervention includes adequate nutritional supplementation, treatment with disease-modifying medications and treatment with hypopnea.
PSG is the gold standard for the diagnosis of sleep-disordered breathing, which can be used to monitor the problem of nocturnal hypopnea in children with SMA. Although expert opinion supports using PSG for diagnosing and noninvasive ventilation to prevent recurrent RRTIs and ARF, there is still no specific standard.2The diagnostic criteria for obstructive sleep apnea are only suitable for patients with obstructive problems in the airway, not for patients with neuromuscular disorders mainly caused by hypopnea.16 The Duchenne muscular dystrophy guidelines17 have proposed the following indication for nocturnal noninvasive ventilation: PCO2> 6.7 kPa (50 mmHg) or > 1.3 kPa (10 mmHg) of the awake baseline at least 2% of the total sleep time, oxygen saturation < 88% for at least five minutes, or AHI > 5 events/h. This standard may be used as a reference for SMA, but the data were not obtained from SMA patients themselves. The questionnaires could be used to assess sleep disorders in SMA.18 but have limitations of subjectivity because nocturnal hypopnea can also be found in pediatric SMA patients without clinical symptoms.19,20 Sleep-disordered breathing is associated with RRTIs in children with SMA, and long-term noninvasive ventilation can improve PSG scores and reduce the incidence of respiratory tract infection.21 Our study used SMA children as samples and analyzed the association between PSG scores and the occurrence of RRTIs and ARF, which ensured the objectivity of the results and applicability to SMA disease. Our study showed that AHI > 10.2 events/h and MSpO2 < 95.5% suggested a high risk for RRTIs and/or ARF in children with SMA.
Lung function declines with age, and the severity is also related to the type of SMA.22,23 Our study also showed that the FVC and PEF % predicted in type 2 were lower than those in type 3 and were lower in the disease group than in the control group. The raw scores of FVC and PEF in type 2 were approximately 60% of the predicted scores, which were approximately 90% in type 3, approximately 40% - 50% in the disease group, and approximately 60% - 70% in the control group. Spirometry can only be performed with the cooperation of children over five years old, so the lung function test could only be used for the evaluation of older children with types 2 and 3. However, for SMA patients, type 1 with younger age is more severe and requires closer observation and proactive intervention. Therefore, for children under five years old, especially those with type 1, it is more important to pay attention to BMIz and PSG scores.
Our study showed that age, MI-E usage and EtCO2 were not associated with RRTIs and/or ARF. The reasons for these inconsistencies with conventional thinking may be as follows. First, the patient’s condition should degenerate with age and progression of the disease. The factor of age should be based on the type and severity of individual cases. Therefore, there was no correlation in our study because the subjects included all three types of SMA. The cutoff points of age to assess the risk for RRTIs and/or ARF for each SMA type might be obtained if the sample size was sufficient. Second, MI-E usage had a positive effect on the intervention against respiratory problems in SMA patients with types 1 and 2.2 However, our study did not show a correlation, indicating that the effect of MI-E usage should be shown only in patients who have a weak cough and need airway clearance techniques. Finally, the diagnostic criterion for sleep-related hypoventilation diseases is PCO2 > 6.7 kPa (50 mmHg), which accounts for more than 25% of the total sleep time according to the international classification of sleep diseases.24 However, our study showed that in 27 cases, only one child of type 2 with severe adenoidal hypertrophy complied with the criteria mentioned above. The OAI of this patient was 29.3 events/h, which was much higher than the HI representing hypopnea of 5.0 events/h. Our study also did not show a correlation between PCO2 and hypoventilation in children with SMA. This might be because the tested value of EtCO2 in patients with neuromuscular disorders was lower than the realistic value.25 On the other hand, this might be consistent with the finding from another study that higher values of PCO2 were not measured in type 1 than in types 2 and 3, and PCO2 levels during sleep might not be used to accurately evaluate hypopnea in children with SMA.3,26
Over the last decade, the approach to treating the respiratory manifestations of SMA has shifted from a reactive approach only when there is a clear indication to a proactive approach of applying therapies earlier in the disease process.2 The use of disease-modifying medications such as nusinersen may delay the decline in lung function, but the effect on long-term improvement is unknown.27-29 Proactive respiratory management and nutrition support still play important roles in the improvement of living conditions in SMA.30,31 Therefore, the assessment of BMIz and PSG should be performed. Furthermore, malnutrition and hypoventilation should be treated to prevent RRTIs and ARF. Our study showed that the accuracy, sensitivity and specificity of the standard of MSpO2 < 96% and AHI > 10 events/h or BMIz < -1 with the occurrence of RRTIs and/or ARF in SMA were 0.798, 0.513 and 0.957, respectively. The high specificity meant that children with SMA who reached this standard would almost certainly develop RRTIs or ARF and should receive intervention.