Article Text
PDF
Abstract
Background Surgery is not a recognised potential amyotrophic lateral sclerosis (ALS) risk factor that might modify the onset or course of ALS.
Methods We studied our database of ALS patients, which includes questions concerning surgical procedures. We defined surgery as an operative procedure requiring general or regional anaesthesia, but not local anaesthesia. Patients were classified as G1—no surgery; G2—surgery performed ≥3 months before disease onset; G3—surgery <3 months before disease onset; and G4—surgery after disease onset. The ALS-FRS score was evaluated every 3 months from presentation. The maximal ALS-FRS score was ascribed to disease onset, itself defined as symptom onset.
Results 657 patients with ALS were studied. In G3 there was a positive correlation between onset-region and surgery-region (p=0.032). In G4, 35 (57.6%) patients had surgery, probably due to initial misdiagnosis. The rate of functional change (%) in G4 was significantly greater in the 3-month period immediately after surgery as compared with the 3-month period before (1.46%±1.35 vs 6.30%±8.10, p=0.005) and the following 3 months (3.30%±3.10, p=0.006).
Conclusions The site of surgery before ALS onset correlates with the region of onset of ALS. Patients with slower disease progression are at an increased risk of undergoing surgery, probably as part of initial difficulty in diagnosis. We noted accelerated disease progression during the 3-month period after surgery. Definite diagnosis is important to avoid unnecessary surgical trauma and subsequent more rapid deterioration.
- Motor Neuron Disease
Statistics from Altmetric.com
Introduction
Several risk factors for amyotrophic lateral sclerosis (ALS) have been described.1 A higher incidence of sporadic ALS (SALS) has been reported in military personnel deployed in the Gulf War,2 and also among athletes, particularly in Italian professional soccer players,3 but the significance of these findings is uncertain. In addition, a relationship between disease onset and trauma has been suggested.4 Nonetheless, most studies of SALS and accidental trauma indicate only a weak association.5 Surgery, which can be regarded as a form of non-accidental trauma, has not generally been considered as a potential risk factor for ALS, although Murros and Fogelholm found no excess past history of surgery in 36 patients in Finland studied between 1981 and 1983.6 We have studied surgery as a possible risk factor determining onset and especially rate of progression of ALS.
Patients and methods
We studied 843 sets of case records from our motor neurone disease (MND/ALS) database, collected from June 1997 to June 2012. All patients with suspected ALS attending our centre completed a questionnaire at their first assessment, which included questions about past trauma (including accidental injuries and surgical interventions), sports and heavy physical activities, infections, autoimmune diseases, and potential toxic exposures. Only patients with progressive weakness during follow-up, fulfilling the modified El Escorial criteria for diagnosis, and with fully completed questionnaires were included in this study. The questionnaire did not change during the period 1997–2013. Patients with familial ALS (FALS) were included, but considered separately, given the likely unique nature of FALS cases. All those included had been followed up at 3-month intervals. Age at disease onset, body mass index (BMI) and disease duration at first clinical evaluation were compared. Disease onset was defined as the first recognition of painless progressive limb or axial weakness without sensory symptoms, of dysarthria or dysphagia, or of progressive respiratory distress. ALS functional rating scale (ALS-FRS) scores were used to evaluate progression.
Surgery was defined as a surgical procedure requiring general or regional anaesthesia. Thus, technically minor interventions such as tonsillectomy, dental procedures and cardiac pacemaker implantation were excluded. Patients were divided into four groups (see figure 1A):
-
Group 1 (G1): patients with no history of surgical intervention in the past.
-
Group 2 (G2): patients who had a surgical procedure more than 3 months before disease onset.
-
Group 3 (G3): patients who had surgery <3 months before symptomatic disease onset.
-
Group 4 (G4): patients who had surgery after disease onset.
The protocol for analysis of ALS patients (A) and change in ALS-FRS during the period of observation ((B) note the acceleration of functional disability in the 3 months after surgery in G4). T0, first symptoms; T1, first visit; T2, second visit; T3, third visit. The grey crosses represent surgery.
The arbitrary period of 3 months (G2 and G3) was determined by the interval between consultations in our centre.
The percentage ALS-FRS change per month was calculated. The maximal ALS-FRS score (40) was ascribed to the time of symptom onset (Time 0; T0), and ALS-FRS values were scored at the first three evaluations at consecutive 3-month intervals: T1 (first observation), T2 and T3 (figure 1A).
Data analysis
We used non-parametric statistics. We used Fisher's exact test to test associations between region of disease onset and site of surgical intervention, according to the following category pairs: bulbar region onset and head or neck surgery; cervical region and upper limb surgery; and lumbar region and lower limb surgery. We considered p<0.05 considered as significant.
Results
Of the total population of 843 ALS patients, we studied 657 (382 men), including 30 with FALS. The 186 patients excluded had incomplete data at entry or follow-up. Mean age at disease onset was 62.6 years (SD 12.1) and disease duration, from first symptom noted at first visit, was 17.1 months (SD 17.8). The disease was of bulbar onset in 193 (29.4%) patients, spinal onset in 443 (67.4%) and respiratory onset in 21 (3.2%). Mean BMI at study entry was 24.9 (SD 3.8). There were no differences regarding age at disease onset (p=0.194), disease duration at first visit (p=0.846) and BMI (p=0.565) between SALS and FALS patients. The 186 excluded patients showed similar gender distribution, disease duration, region of onset, age at disease onset and BMI, compared with the studied group.
Of the 657 patients, 464 (70.6%) had no history of surgery (G1) and 111 (16.9%) had at least one eligible surgical intervention more than 3 months before disease onset (G2)—mean time 14.6 years (SD 16.0). In all, 24 of the FALS patients were in G1 (80%) and 4 (13%) in G2. A total of 23 patients (3.5%), all with SALS, were included in G3 and 59 (9%) patients (two with FALS, 7% of the FALS patients) in G4. No significant differences between the groups were found regarding gender (p=0.873). There were no significant differences in disease duration among G1 (16.6±18.2 months), G2 (16.8±18.6 months) and G3 (16.4±13.9 months) (p>0.7). Patients in G4 had a longer duration of symptomatic ALS (21.7±13.9 months) when compared with G1 (p<0.001), G2 (p<0.001) and G3 (p=0.015). There were no statistical differences between groups regarding BMI (p>0.05). Patients in G1 and G4 were significantly younger at disease onset than patients in G2 (p=0.001 and p=0.007, respectively) but not as compared with G3. In the FALS group, there were no differences regarding age at disease onset and disease duration at first visit with or without a history of surgery (p>0.5). Men with spinal-onset SALS, without a history of surgery, were younger than those with a history of surgery before disease onset (p<0.02). In 35 of the 57 patients included in G4, surgical intervention probably derived from misdiagnosis (see online supplementary table); a positive history of ALS was not protective against a wrong surgical procedure (5% for sporadic cases, 7% for FALS). All surgical interventions are listed in the online supplementary file.
In G3, localisation of surgery and region of ALS onset were correlated (p=0.032). All 15 spinal onset patients had surgery in a spinal region, but five of the eight bulbar onset patients had surgery in a spinal region, rather than to the head and neck region. This correlation probably reflects misdiagnosis of a surgical spinal condition.
Surgery and ALS progression
In only 15 out of the 57 SALS patients and in two FALS patients (G4), 17 patients in total, who underwent surgery less than 4 weeks before T1, the ALS-FRS score at the time of surgery was considered as that defined at T1. In these 17 patients, the per cent ALS-FRS score change per month was significantly greater within the 3-month period after surgery, as compared before (6.3%±8.1 vs 1.46%±1.35, p=0.005) and in the following period of 3 months (6.3%±8.1 vs 3.30%±3.10, p=0.006). In the remaining patients in G4, operated more than 4 weeks before T1, we were inevitably unable to estimate the ALS-FRS score at the time of surgery. For the whole group of G1, G2 and G3 patients, the per cent change in ALS-FRS score was less between T0 and T1 than in T1 and T2, but significantly increased in T2 and T3, as opposed to G4 in which the most rapid change in ALS-FRS score occurred after surgery (T1 and T2). No statistical difference was found between G4 and non-G4 patients regarding the percentage of ALS change between T0 and T3 (p>0.843; table 1). The effect of surgery in accelerating functional decline in ALS is shown in figure 1B. In G4, ALS-FRS score declined a mean value of 7.2 between T1 and T2, twice as compared with the rest of the patients (mean of 3.3), but within the T2 and T3 interval the mean decrement in G4 was 3.8, which is similar to the remaining patients (mean of 2.6).
Statistical analysis, showing significant results (p<0.01) in bold in last two columns
Discussion
The onset of ALS is notoriously difficult to define. Although some investigators believe there may be a long latent interval before weakness develops there is currently little evidence to support this concept. For example, a clinical and MUNE study of a SOD-1 family revealed an abrupt onset of reduced MUNE and then weakness, with subsequent asymmetrical progression.7 Overall, about 20% of ALS patients had an eligible surgical intervention before symptom onset, a risk of 0.5%/year for surgical intervention during adult life for these patients. The national rate of major surgery/year for the Portuguese population was 6.6% in 2005.8 This national statistic does not note multiple procedures, or provide an age-related breakdown, and includes minor procedures.8 Our observation that younger ALS patients with slower disease progression had an increased incidence of surgery after disease onset was likely to be related to initial misdiagnosis (56%; see online supplementary file).
There was a smaller per cent change in T0–T1 ALS-FRS scores when compared with T1–T2 in all groups suggesting awareness of early ALS symptoms. Contrary to Kollewe and collegues,9 it is likely that ALS-FRS change is slower in the early phases of ALS. In all groups, except for G4, disease progression appeared more rapid in T2–T3 than in T1–T2 (table 1), suggesting an immediate, transitory, negative effect of surgical intervention on the rate of progression in G4; although, the ordinal nature of the ALS-FRS scale should be remembered.
There are many possible reasons why surgical interventions could accelerate progression of ALS. Anaesthetic drugs themselves could be involved as they modify calcium homeostasis, membrane receptors and ionic channel function, as well as neurotransmitter activity. In addition, anaesthetic agents may cause suppression of neurogenesis, hypoperfusion, changes in metabolism, inflammatory stimulation and apoptosis.10–12 Surgical stress causes systemic inflammation, increases the risk of central nervous system microembolism, increases cortisol release and disturbs the sleep pattern. There are earlier suggestions that surgical stress and anaesthesia may influence neurodegenerative processes.13–15 Indeed, biomarkers of neuronal damage, such as neurone-specific enolase, S100B, nuclear factor kB, hypophosphorylated neurofilament H and deubiquitinating enzyme UCH-L1, and of inflammation, such as TNF-a, IL-1b and IL-6, have been found to be increased after surgery.14 ,15
Potential problems with our study include reliance on retrospective analysis of data, the number of patients included in G4 is relatively small and the patients were submitted to very different types of surgical procedures. Nonetheless, the processes determining the rate of progression of ALS are not currently understood;1 indeed, this issue has scarcely been addressed in the contemporary literature.
Acknowledgments
This work was supported by national funds through Fundação para a Ciência e a Tecnologia (FCT), under project contract PTDC/EIAEIA/111239/2009—Neuroclinomics.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Data supplement 1 - Online table
Footnotes
-
Contributors All authors made a substantial contribution to the manuscript. SP organised and analysed the data, and prepared a draft of the manuscript; MdC collected the data over the period of time in which the patients were observed, interpreted the final data a prepared the manuscript; MS contributed in the study conception, data interpretation and manuscript preparation.
-
Competing interests None.
-
Funding Provided grant to support an investigator.
-
Ethics approval Local Ethics Committee.
-
Provenance and peer review Not commissioned; externally peer reviewed.
-
Data sharing statement The data contained in this work are included in a large database of ALS patients followed in Lisbon. These data are not available for other investigators, except for joined projects and after previous approval of the Ethics committee.
Linked Articles
- Editorial commentary