Amyotrophic lateral sclerosis (ALS) is a form of degenerative motor neuron disease of unknown etiology. The disease is characterized by progressive muscle weakness and atrophy throughout the body 1 2 3 . The prevalence is 4–8 in 100 000 and the annual incidence is 1–2 in 100 000 4 . Prognosis is poor with a median survival from the onset of symptoms of 2–4 years 5 6 . Despite extensive effort, no curative treatment is available and riluzole (a tetrodotoxin-sensitive sodium channel blocker) is the only drug that can slow the progression of the disease 7 8 . Therefore, treatment following the diagnosis is palliative 9 10 .

Hypoventilation due to respiratory insufficiency is the most common cause of death in patients with advanced ALS 3 5 . Non-invasive ventilation (NIV) has been recommended for ALS patients when hypoventilation occurs because it relieves dyspnoea, increases the quality of life, and may prolong survival in late stage ALS patients 11 12 13 . The current guidelines recommend beginning NIV if the patient presents with dyspnoea, orthopnoea, disturbed sleep, tachypnoea, nocturnal desaturation < 90%, increased morning carbon dioxide partial pressure (pCO₂) > 6 kPa, decreased sniff nasal pressure < 40 cmH₂O, decreased maximum inspiratory mouth pressure (MIP) < 60 cmH₂O, or decreased forced vital capacity (FVC) < 80% 9 10 14 . However, there is great international variation in the use of different diagnostic tests prior to initiation of NIV.

The onset of muscle weakness varies between patients and symptoms may appear first in the limbs or they may start from the bulbar area leading to dysphagia and speech difficulties 3 . This wide variability in the clinical course of ALS can be challenging for the clinician because the progression of respiratory insufficiency also greatly varies among these patients 1 3 . Because of this variability, the current guidelines recommend a clinical visit every 2–3 months 9 10 . Although nocturnal desaturation and carbon dioxide tension can be recorded noninvasively, it is often not possible to perform these measurements outside the hospital because of lack of devices and health care providers to assist the patient. The measurement of increased morning pCO₂ requires blood sampling by a health care professional. In addition, spirometric measurements require good facial function, as patients have to hold their lips tightly around the mouthpiece of the spirometer. In ALS patients with severe bulbar dysfunction, this measurement may be unreliable 10 15 . We therefore wished to determine if we could use respiratory rate and thoracic movement as preliminary clinical signs in the development of respiratory insufficiency.

The records for 42 patients with ALS (20 male, 22 female) were adequate for analysis. The characteristics of the patients are shown in Table 1. The age and the duration of the symptoms before the diagnosis of ALS did not differ between the groups.

The difference between the groups was assessed using the Mann Whitney U test.

NIV = non-invasive ventilation.

The mean pCO₂ and pO₂ in the morning arterial blood gas samples at the time of diagnosis in Group 1 were 6.3 (5.6–6.9) and 9.8 (9.3–10.4) kPa, respectively. The mean pCO₂ and pO₂ at the time of diagnosis in Group 2 were 5.3 (5.0–5.6) and 10.7 (9.4–12) kPa, respectively. The mean pCO₂ was significantly higher in Group 1 (p = 0.007) whereas the values for mean pO₂ did not differ between the two groups (p = 0.4). The mean respiratory rate at the time of diagnosis in Group 1 was 21 (18–24) BPM and 16 (14–18) BPM in Group 2 (p = 0.005). The mean thoracic movement at the time of diagnosis in Group 1 was 2.9 (2.2–3.6) cm and 4.0 (3.4–4.8) cm in Group 2 (p = 0.01). We observed no other differences between the two groups (Table 2).

*Determined using the Mann Whitney U test.

n = number of patients completing the measurement; CI = confidence interval; FVC = forced vital capacity expressed as % from the reference values; FEV 1 = forced expiratory volume exhaled in one second and expressed as % from the reference values; MIP = maximal inspiratory mouth pressure; MEP = maximal expiratory mouth pressure; SNP = sniff nasal pressure; PCF = peak cough flow. All measurements were performed three times in a sitting position and the best values were recorded.

The severity of hypoventilation symptoms was assessed using a visual analogue scale (VAS) in which 0 represented no symptoms and 10 represented the worst symptoms that the patients could imagine. Thoracic movement was measured using a measuring tape and recorded as the difference in the thoracic circumference during maximal inspiration and expiration at the mammillary level.

This was a retrospective study to clarify if we could use the respiratory rate and thoracic movement as preliminary clinical signs for the development of significant respiratory insufficiency in ALS patients. The main finding in our study was that patients who received NIV within six months of the diagnosis of ALS had higher respiratory rates and smaller thoracic movement at diagnosis compared with patients who received NIV later.

Among healthy subjects, a respiratory rate under 15 BPM allows a physiologically optimal level of work of breathing 18 19 . In advanced ALS, thoracic compliance is usually decreased, resulting in decreased tidal volume 20 , increased respiratory rate and work of breathing 18 19 . NIV can help patients compensate for the decreased thoracic compliance and thus decrease the respiratory frequency and the work of breathing 20 21 . Although the retrospective study design and small number of patients did not allow the calculation of positive and negative predictive values and cut-off values for respiratory rate and thoracic movement in the assessment of respiratory insufficiency in ALS, it appears that NIV was initiated within six months from diagnosis for those patients who had a respiratory rate at least of 20 BPM at the moment of diagnosis.

Another result of interest was that we failed to show any difference in dyspnoea between the two study groups (Table 2). The patients scored only mild symptoms of dyspnoea and morning headaches at the moment of diagnosis in both groups. The same was true for sleep disturbances and daytime sleepiness. In a previous study with 36 ALS patients, respiratory complaints did not occur until the vital capacity, MIP, or MEP were severely impaired 22 . In that study, the strongest correlation with dyspnoea was a decline in vital capacity. The authors suggested that a small reduction in vital capacity does not cause any symptoms except in heavy exercise. The patients decrease their physical activity because of skeletal muscle weakness, and thus may not feel dyspnoeic during their daily activities unless more strenuous exercise is required. Our results are similar as our patients complained of very little dyspnoea at the time of diagnosis. Nevertheless, as our patients were aware of the possibility of initiating NIV, they might have been reluctant to disclose the severity of their symptoms and intentionally underestimated their response in our VAS query. A larger sample size might show a difference in dyspnoea.

FVC and FEV 1 are commonly used to assess pulmonary function in lung diseases 23 and in ALS patients. Previously, FVC, SNP, MIP and MEP were determined in 16 ALS patients monthly over a period of 18 ± 10 months 24 . The SNP, MIP and MEP were severely reduced even though the FVC remained normal. In addition, the measurement of MIP and MEP was difficult in these advanced ALS patients because they had difficulties tightly holding the mouthpiece of the recording device. This led to air leaks and reduced the values of MIP and MEP. A reduced SNP of under 40 cmH₂O has been suggested as the most sensitive and easiest test to perform in ALS patients 9 10 24 . However, although we found that the SNP was severely reduced in both groups, only those with an increased respiratory rate in Group 1 were ready to start NIV (Table 2).

Our study has limitations. First, the retrospective design compared the respiratory measurements between two different patient groups at one single time point. Thus, we cannot draw definitive conclusions for the cut-off values for respiratory rate or thoracic movement to indicate the need for NIV. Further studies are needed. In addition, the only inclusion criterion was the respiratory assessment within three months from diagnosis. Thus, we did not evaluate other pulmonary or heart diseases which might have biased the results. Second, the ideal is a randomized prospective trial in which the initiation of NIV is blinded and the subjects are observed from an asymptomatic stage until severe hypoventilation occurs. However, because NIV is established in palliative care in ALS 9 10 , ethical aspects have to be considered in randomized study designs, especially if the initiation of NIV is somehow blinded. Moreover, the diagnosis of ALS is often delayed and the patients often have reduced ventilation capacity at the time of diagnosis 22 25 . We saw this in our patients as the median time from the first symptoms until diagnosis ranged from 11 to 13 months (Table 1). During this time, the patients’ symptoms were evaluated at the primary care level or not at all. Thus, when we met the patients, half already showed moderately to severely reduced FVC (60%), SNP (24 cmH₂O), MIP (29 cmH₂O), or MEP (41 cmH₂O) and NIV was recommended based on a single measurement without follow-up. Third, we dealt with severely ill patients, whose NIV initiation was based on both the clinical measurements and the patient’s own desire. The fact that NIV was based on the patient’s own desire may have biased the answers in the hypoventilation questionnaire if the patients reported no or only mild symptoms of hypoventilation even if they had severely reduced respiratory measurements.

The use of NIV as palliative treatment has previously been shown to relieve symptoms as well as to improve survival in ALS patients 11 12 13 . However, the optimal timing for initiating NIV is not yet well established. Most of the available methods used to evaluate the degree of hypoventilation and the need for NIV require well-trained health care professional and special devices. In contrast, assessing resting respiratory rate or thoracic movement can be performed in a patient’s home by primary care nursing staff and could be a feasible screening tool for detecting hypoventilation. However, further studies are needed to define the positive and negative predictive values and cut-off values for respiratory rate and thoracic movement in the assessment of respiratory insufficiency in ALS.

ALS: Amyotrophic lateral sclerosis; NIV: Non-invasive ventilation; FVC: Forced vital capacity; FEV 1: Forced expiratory volume exhaled in 1s; PCF: Peak cough flow; MIP: Maximum inspiratory mouth pressure; MEP: Maximum expiratory mouth pressure; SNP: Sniff nasal pressure.

The authors declare that they have no competing interests.

WS, TS, AV, KO and RA designed the study. WS, TS and AV collected the data. WS and TS were responsible for the interpretation of the data. WS, TS and SS prepared the manuscript. All authors participated in critical revision of the article for important intellectual content, and approved the final version of the article.